Multimodal remote sensing data classification can enhance the model’s ability to distinguish land features through multimodal data fusion. In this context, how to help models understand the relationship between multimodal data and target tasks is the focus of researchers. Inspired by human feedback learning mechanisms, causal reasoning mechanisms, and knowledge induction mechanisms, this paper integrates causal learning, reinforcement learning, and meta learning into a unified remote sensing data classification framework and proposed the causal meta-reinforcement learning (CMRL). First, based on feedback learning mechanisms, we have overcome the limitations of traditional implicit optimization of fusion features and customized a reinforcement learning environment for multimodal remote sensing data classification tasks. Through feedback interactive learning between agents and the environment, we help them understand the complex relationships between multimodal data and labels, thereby achieving full mining of multimodal complementary information. Second, based on the causal inference mechanism we designed causal distribution prediction action, classification reward, and causal intervention reward, capturing pure causal factors in multimodal data and cutting off false statistical associations between non-causal factors and class labels. Finally, based on the knowledge induction mechanism, we designed a bi-layer optimization mechanism based on meta-learning. By constructing a meta training task and meta validation task simulation model in the generalization scenario of unseen data, we helped the model induce cross-task shared knowledge, thereby improving its generalization ability for unseen multimodal data. The experimental results on multiple sets of multimodal datasets show that the proposed method achieved state-of-the-art performance in multimodal remote sensing data classification tasks.

Ensuring precise segmentation of colorectal polyps holds critical importance in the early diagnosis and treatment of colorectal cancer. Nevertheless, existing deep learning-based segmentation methods are fully-supervised, requiring extensive precise manual pixel-level annotation data, which leads to high annotation costs. Additionally, it remains challenging to train large-scale segmentation models when confronted with limited colonoscopy data. To address these issues, we introduce the general segmentation foundation model—Segment Anything Model (SAM) into the field of medical image segmentation. Fine-tuning the foundation model is an effective approach to tackle sample scarcity. However, current SAM fine-tuning techniques still rely on precise annotations. To overcome this limitation, we propose WSPoly-SAM, a novel weakly-supervised approach for colonoscopy polyp segmentation. WSPoly-SAM utilizes weak annotations to guide SAM in generating segmentation masks, which are then treated as pseudo-labels to guide the fine-tuning of SAM, thereby reducing the dependence on precise annotations data. To improve the reliability and accuracy of pseudo-labels, we have designed a series of enhancement strategies to improve the quality of pseudo-labels and mitigate the negative impact of low-quality pseudo-labels. Experimental results on five medical image datasets demonstrate that WSPoly-SAM outperforms current fully-supervised mainstream polyp segmentation networks on the Kvasir-SEG, ColonDB, CVC-300, and ETIS datasets. Furthermore, by using different amounts of training data in weakly-supervised and fully-supervised experiments, it is found that weakly-supervised fine-tuning can save 70% to 73% of annotation time costs compared to fully-supervised fine-tuning. This study provides a new perspective on the combination of weakly-supervised learning and SAM models, significantly reducing annotation time and offering insights for further development in the field of colonoscopy polyp segmentation.

To further enhance the intelligence level of coal mining faces and achieve autonomous derivation, learning, and optimization of shearer navigation cutting, this paper proposes the methods of shearer digital twin navigation cutting motion planning based on the concept of shearer autonomous navigation cutting technology and intelligent coal mining face digital twins. The study includes the digital twin theory and the construction method of the shearer digital twin navigation cutting motion planning system based on this theory. Based on the digital twin theory, a shearer digital twin navigation cutting motion planning system was constructed. This system supports the service functions of shearer cutting digital twin, dynamic navigation map digital twin, reinforcement learning environment construction, and motion planning through the physical perception layer, comprehensive data layer, and digital-model fusion analysis layer. Finally, by comparing the effects of the DQN-NAF and DDPG deep reinforcement learning algorithms in the shearer motion planning task within the constructed digital twin environment, the results show that the DQN-NAF algorithm demonstrates better performance and stability in solving the shearer digital twin motion planning task.

As split learning (SL) becomes popular for addressing data privacy and resource limitation issues in edge model training process, it is crucial to pre-validate whether the extensive use of wireless and computing resources in SL can achieve the expected training latency and convergence. While digital twin network (DTN) can potentially estimate network strategy performance in pre-validation environments, they are still in their infancy for SL tasks, facing challenges like unknown non-i.i.d. data distributions, inaccurate channel states, and misreport resource availability across devices. To address these challenges, this paper proposes a TransNeural algorithm for DTN pre-validation environment to estimate SL latency and convergence. First, the TransNeural algorithm integrates transformers to efficiently model data similarities between different devices, considering different data distributions and device participate sequence greatly influence SL training convergence. Second, it leverages neural network to automatically establish the complex relationships between SL latency and convergence with data distributions, wireless and computing resources, dataset sizes, and training iterations. Deviations in user reports are also accounted for in the estimation process. Simulations show that the TransNeural algorithm improves latency estimation accuracy by 13.44% and convergence estimation accuracy by 116.74% compared to traditional equation-based methods.

Particulate organic carbon (POC) is a pivotal component within the marine carbon cycle, actively involved in diverse biogeochemical processes. This study focuses on Mediterranean as the research region, selecting the most influential factors from 47 potential ones influencing surface POC. Using geographic detector for factor identification, ten primary influencers were identified. Preliminarily optimized random forest (BRF), backpropagation neural network, adaptive boosting, and extreme gradient boosting were utilized to construct POC assessment models. Notably BRF exhibited superior performance in estimating the sea surface POC. Further optimization of RF using the tuneRanger R package resulted in R² of 0.868, mean squared error of 1.119 (mg/m³) ², and mean absolute error of 1.041 (mg/m³). Surface POC concentrations in the Mediterranean for May and June 2017 were estimated. Spatial distribution analysis unveiled higher concentrations in the west, lower in the east, higher in the north, lower in the south, with higher levels near the coast and lower far from the coast. Additionally, the study deliberated on the impact of human activities on surface POC in the Mediterranean. This research contributes a high-precision method for satellite retrieval of surface POC concentrations in the Mediterranean, thereby enriching the understanding of POC dynamics in the area.

Heart failure is a prevalent cardiovascular condition with significant health implications, necessitating effective diagnostic strategies for timely intervention. This study explores the potential of continuous monitoring of non-invasive signals, specifically integrating Photoplethysmogram (PPG) and Electrocardiogram (ECG), for enhancing early detection and diagnosis of heart failure. Leveraging a dataset from the MIMIC-III database, encompassing 682 heart failure patients and 954 controls. Feature selection techniques were used to systematically select key features which were identified for their clinical relevance and significance in capturing cardiovascular dynamics and to reduce computational complexity and to decrease the chance of overfitting the ML algorithms. These features are then utilized to train and evaluate machine learning algorithms, resulting in a model with an impressive accuracy of 98%, sensitivity of 97.60%, specificity of 96.90%, and precision of 97.20%. The integrated approach outperforms single-signal strategies, showcas-ng its potential for early, precise, and non-invasive heart failure diagnosis. Furthermore, the study underscores the significance of continuous monitoring through wearables, emphasizing the benefits of integrating multiple signals for a comprehensive evaluation of cardiovascular health. The proposed approach holds promise for implementation in hardware systems to enable continuous monitoring, aiding in early detection and prevention of critical health conditions.

Microscopic and ultra-microscopic vascular sutures are indispensable in the surgical procedures such as arm transplantation and finger reattachment. The state of the blood vessels after suturing, such as vascular patency, narrow, and blocked, determines the success rate of the operation. If we can grasp the golden window period after blood vessel suture and before muscle tissue suture to achieve accurate and objective assessment of blood vessel status, it can not only reduce medical costs but also improve social benefits. Doppler optical coherence tomography enables high-speed, high-resolution imaging of biological tissues, especially microscopic and ultra-microscopic blood vessels. Using Doppler optical coherence tomography to image the sutured blood vessels, not only the three-dimensional structure of the blood vessels, but also the blood flow information can be obtained. By extracting the contour of blood vessel wall and the contour of blood flow area, the three-dimensional shape of blood vessel can be reconstructed in three dimensions, which provides parameter support for the assessment of blood vessel status. In this work, we propose a neural network-based multi-classification deep learning model that can simultaneously and automatically extract blood vessel boundaries from Doppler OCT vessel intensity images and the contours of blood flow regions from corresponding Doppler OCT vessel phase images. Compared with the traditional random walk segmentation algorithm and cascade neural network method, the proposed model has better performance. This method is easier to realize system integration and has great potential for clinical evaluation. It is expected to be applied to the evaluation of microscopic and ultra-microscopic vascular status in microvascular anastomosis.

(1) Background: We aim to construct a machine learning (ML) algorithm to predict the risk of distant metastasis (DM) of T1 and T2gallbladder cancer (GBC); (2) Demographic and clinical pathological data of T1 and T2 GBC patients were extracted from the National Institutes of Health (NIH)’s Surveillance, Epidemiology, and End Results (SEER) database between 2004 and 2015 to develop seven ML algorithm models. Models were evaluated based on accuracy, precision, recall rate, F1- score, and area under the receiver operating characteristic curve (AUC); (3) Results:A total of 4371 patients were included in the study, of whom 764 (17.4%) developed DM. Multivariate logistic regression showed that age, histology, tumor size, T and N stages were independent factors in GBC with DM. A novel nomogram was established to predict distant metastasis in early T stage GBC patients. Evaluation indicators of the best model Random Forest (RF) were as follows: accuracy (0.828), recall rate (0.862), precision (0.811), F1- score (0.836), and AUC value (0.913); (4) Con-clusions: The RF model constructed in this study could accurately predict distant metastasis in GBC patients, which may provide clinicians with more personalized clinical decision-making recom-mendations.

Accurately estimating shrub biomass in arid and semi-arid regions is critical for understanding ecosystem productivity and carbon stocks at both local and global scales. However, capturing the shrub biomass accurately by satellite observations is challenging due to the short and sparse features of shrubs. In this study, we presented a framework to estimate shrub biomass at a 10m spatial resolution by integrating ground in-situ data, unmanned aerial vehicle (UAV), Sentinel-2, and Sentinel-1 observations. The pilot study was conducted in the Helan mountains of Ningxia province, China. First, the spatial distribution map of shrubland and other land cover types was generated in 2023. Then, a prediction model of shrub biomass was developed in a Random Forest Regression (RFR) approach driven by different predicted variable datasets based on in-situ measurements, UAV, and satellite images. Finally, the developed model was used to produce the biomass map of shrubland in 2023. The uncertainty was characterized by creating a standard deviation (SD) map using the Leave-One-Out Cross-Validation (LOOCV) method in the shrub biomass estimates. The resultant shrubland distribution map in 2023 has an F1 score of 0.92. The shrub biomass model driven by spectral bands and vegetation indices (R2= 0.62) demonstrated superior performance than that only driven by spectral bands (R2=0.33) or vegetation indices (R2=0.55). The uncertainty of the estimated biomass was lower than 4%, with the lowest values (<2%) happening in the high shrub coverage (>30%) and biomass production (>30kg/m2) regions. Furthermore, the findings show that the magnitude of shrub biomass was affected by the aridity and precipitation significantly in the study region. This study provides a workflow to accurately monitor the biomass of shrublands in a complicated environment based on remote sensing images from multi-platform observations.

The incessant evolution of online platforms has ushered in a multitude of shopping modalities. Within the food industry, however, assessing the delectability of meals can only be tentatively determined based on consumer feedback encompassing aspects such as taste, pricing, packaging, service quality, delivery timeliness, hygiene standards, and environmental considerations. Traditional text data mining techniques primarily focus on consumers' emotional traits, disregarding pertinent information pertaining to the online products themselves. In light of these aforementioned issues in current research methodologies, this paper introduces the Bert BiGRU Softmax model combined with multimodal features to enhance the efficacy of sentiment classification in data analysis. Comparative experiments conducted using existing data demonstrate that the accuracy rate of the model employed in this study reaches 90.9%. In comparison to single models or combinations of three models with the highest accuracy rate of 7.7%, the proposed model exhibits superior accuracy and proves highly applicable to online reviews.

Incorporating knowledge graphs as auxiliary information to enhance recommendation systems can improve the representations learning of users and items. Recommendation methods based on knowledge graph can introduce user-item interaction learning into the item graph, focusing only on learning the node vector representations within a single graph; alternatively, they can treat user-item interactions and item graphs as two separate graphs and learn from each graph individually. Learning from two graphs has natural advantages in exploring original information and interaction information, but faces two main challenges: 1) in complex graph connection scenarios, how to adequately mine the self-information of each graph, and 2) how to merge interaction information from the two graphs while ensuring that user-item interaction information predominates. Existing methods do not thoroughly explore the simultaneous mining of self-information from both graphs and effective interaction information, leading to the loss of valuable insights. Considering the success of contrastive learning in mining self-information and auxiliary information, this paper proposes a dual-graph contrastive learning recommendation method based on knowledge graphs (KGDC) to explore a more accurate representations of users and items in recommendation systems based on external knowledge graphs. In the learning process within the self-graph, KGDC has strengthened and represented the information of different connecting edges in both graphs, and extracted the existing information more fully. In interactive information learning, KGDC reinforces the interaction relationship between users and items in the external knowledge graph, realizing the leading role of the main task. We have conducted a series of experiments on three standard datasets, and the results show that the proposed method can achieve better results.

Accurate crop diseases classification is crucial for ensuring food security and enhancing agricultural productivity. However, existing crop disease classification algorithms primarily focus on a single image modality and typically require a large number of samples. Our research counters these issues by using pre-trained Vision-Language Models (VLMs), which enhances multimodal synergy for better crop disease classification than traditional unimodal approaches. Firstly, we apply the multimodal model Qwen-VL to generate meticulous textual descriptions for representative disease images selected through clustering from the training set, which will serve as prompt text for generating classifier weights. Compared to using solely the language model for prompt text generation, this approach better captures and conveys fine-grained and image-specific information, thereby enhancing prompt quality. Secondly, we integrate the cross-attention and the SE (Squeeze-and-Excitation) attention into the training-free mode VLCD (Vision-Language model for Crop Diseases classification) and the training-required mode VLCD-T (VLCD-Training) respectively for prompt text processing, enhancing classifier weights by emphasizing key text features. Experimental outcomes conclusively prove our method’s heightened classification effectiveness in few-shot crop disease scenarios, tackling data limitations and intricate disease recognition issues. It offers a pragmatic tool for agricultural pathology and reinforces the smart farming surveillance infrastructure.

Scene classification plays an important role in the intelligent processing of high-resolution satellite (HRS) remotely sensed image. In HRS image classification, multiple features, e.g. shape, color, and texture features, are employed to represent scenes from different perspectives. Accordingly, effective integration of multiple features always results in better performance compared to methods based on a single feature in the interpretation of HRS image. In this paper, we introduce a multi-task joint sparse and low-rank representation model to combine the strength of multiple features for HRS image interpretation. Specifically, a multi-task learning formulation is applied to simultaneously consider sparse and low-rank structure across multiple tasks. The proposed model is optimized as a non-smooth convex optimization problem using an accelerated proximal gradient method. Experiments on two public scene classification datasets demonstrate that the proposed method achieves remarkable performance and improves upon the state-of-art methods in respective applications.

Within environmental perception, automatic navigation and object detection, computer vision is a crucial and demanding field with many applications in modern industries, such as multi-target long-term visual tracking in automated production, defect detection, and driverless robotic vehicles with sensing ability. The performance of computer vision has greatly improved recently thanks to developments in deep learning algorithms and hardware computing capabilities, which has spawned the creation of a large number of related applications. This paper presents the results of a detailed review of over 50 papers published over the course of two decades (1999–2024), with a primary focus on the technical advancement of computer vision. To elucidate the foundational principles, an examination of typical visual algorithms based on traditional correlation filtering was initially conducted. Subsequently, a comprehensive overview of the most recent advancements in deep learning-based computer vision techniques was compiled. Furthermore, a comparative analysis between conventional and novel algorithms was undertaken to discuss the future trends and directions of computer vision. Lastly, the feasibility of employing visual SLAM (Simultaneous Localization and Mapping) algorithms in the context of autonomous vehicles with sensing capability was explored. Furthermore, we explored a thorough and efficient architecture for utilizing these techniques in autonomous robotic vehicles into unmanned green and low-carbon smart factories, underlining novel advances as well as potential prospects for future research.

Deep learning has increasingly been employed to identify minerals. However, deep learning can only be used to identify minerals in the distribution of the training set, while any mineral outside the spectrum of the training set is inevitably categorized erroneously within a predetermined class from the training set. To solve this problem, the study introduces the approach that amalgamates One-Class Support Vector Machines (OCSVM) with the ResNet architecture for the out-of-distribution mineral detection. Initially, ResNet undergoes training using a training set comprising well-defined minerals. Subsequently, the earlier layers of the trained ResNet are employed to extract the discriminative features of the mineral under consideration. These extracted mineral features then become the input for OCSVM. When OCSVM discerns the mineral in the training set's distribution, it triggers the subsequent layers within the trained ResNet, facilitating the accurate classification of the mineral into one of the predefined categories encompassing the known minerals. In the event OCSVM identifies the mineral out of the training set's distribution, it is unequivocally categorized as an unclassified or 'unknown' mineral. Empirical results substantiate the method's capability to identify out-of-distribution minerals while concurrently maintaining a commendably high accuracy rate for the classification of the 36 in-distribution minerals.

High-power laser facilities necessitate predicting incremental damage to final optics to identify evolving damage trends. In this study, we propose a surface damage detection method utilizing image segmentation employing ResNet-18, and a damage area estimation network employing U-Net++. Paired sets of online and offline images of optics obtained from a large laser facility is used to train the network. The trends of varying damage could be identified by incorporating additional experimental parameters. A key advantage of the proposed method is that the network can be trained end-to-end on small samples, eliminating the need for manual labeling or feature extraction. The software developed based on these models can facilitate the daily inspection and maintenance of optics in large laser facilities.

Recommendation systems are crucial in navigating the vast digital market. However, user data’s dynamic and non-stationary nature often hinders their efficacy. Traditional models struggle to adapt to the evolving preferences and behaviours inherent in user interaction data, posing a significant challenge for accurate prediction and personalisation. Addressing this, we propose a novel theoretical framework, the Non-stationary Transformer, designed to capture and leverage the temporal dynamics within data effectively. This approach enhances the traditional transformer architecture by introducing mechanisms accounting for non-stationary elements, offering a robust and adaptable solution for recommendation systems. Our experimental analysis, encompassing deep learning and reinforcement learning paradigms, demonstrates the framework’s superiority over benchmark models. The empirical results confirm the efficacy of our proposed framework, which not only provides significant performance enhancements, approximately 8% in Logloss reduction and up to 2% increase in F1 score but also underscores its potential applicability across accumulative reward scenarios. These findings advocate adopting Non-stationary Transformer models to tackle the complexities of today’s recommendation tasks.

Named Entity Recognition (NER) serves as a fundamental and pivotal stage in the development of various knowledge-based support systems, including knowledge retrieval and question answering systems. In the domain of pig diseases, Chinese NER models encounter several challenges such as the scarcity of annotated data, domain-specific vocabulary, diverse entity categories, and ambiguous entity boundaries.To address these challenges, a corpus, labeled datasets and a lexicon specific to pig diseases for Chinese named entity recognition were constructed. Subsequently, a Pig Disease Chinese Named Entity Recognition (PDCNER) model was proposed. The model integrates external lexicon knowledge of pig disease by employing Lexicon-enhanced BERT and enhance feature representation by incorporating contrastive learning. Experimental results show that the model achieved the best recognition results, with a precision of 86.92%, a recall of 85.08%, and an F1-score of 85.99% respectively. Furthermore, the model exhibits robustness and generalizability across few-shot and publicly available datasets. Experimental results illustrate the proposed model could effectively identify Chinese named entities of pig diseases, outperforming several existing baseline methods. Moreover, the proposed model can be extended to other animal disease domains, such as chicken and cattle, thereby facilitating seamless adaptation for named entity identification across diverse contexts.

Large-scale deep learning models have produced significant results in underwater visual image target detection. However, deploying these models on underwater embedded devices poses a series of issues and challenges. Firstly, underwater devices have limited communication capabilities and carry relatively short supply of energy, resulting in constrained computing power for edge devices. Secondly, the large scale of deep neural networks creates a conflict in power demands with unmanned underwater equipment. Thirdly, insufficient underwater light and turbid water quality will lead to degradation problems in the images acquired by the device. In this paper, we build an offline system to complete the task of underwater scene target recognition. This paper introduces several innovations: 1) The new YOLO-TN model is developed by compressing the target recognition model based on YOLO-V5. The model use a model compression algorithm replace the YOLO-V5’s backbone network with a lightweight network, then perform parameter pruning on the model detection head, while reduce the size of the input image. The model uses a Teacher-guided Neural Architecture Search method based on YOLO-V5, named as YOLO-TN. 2)A method for constructing and processing real underwater environment data sets is designed. The datasets are all captured using underwater unmanned vehicles in real underwater environments, addressing issues present in existing underwater datasets, such as unbalanced target volume and single-image environment. Experimental results demonstrate that the YOLO-TN model significantly reduces the model size while maintaining the original recognition accuracy. Furthermore, it achieves 28.6 FPS on embedded devices, which is 12 times that of the original model.

Hyperspectral images (HSI) contain abundant spectral information. Efficient extraction and utilization of this information for image classification remain prominent research topics. Previously, hyperspectral classification techniques primarily relied on statistical attributes and mathematical models of spectral data. Deep learning classification techniques have recently been extensively utilized for hyperspectral data classification, yielding promising outcomes. This study proposes a deep learning approach that uses polarization feature maps for classification. Initially, the polar coordinate transformation method is employed to convert the spectral information of all pixels in the image into spectral feature maps. Subsequently, the proposed Deep Context Feature Fusion Network (DCFF-NET) is utilized to classify these feature maps. The model is validated using three open-source hyperspectral datasets: Indian Pines, Pavia University, and Salinas. The experimental results indicated that DCFF-NET achieves excellent classification performance. Experimental results on three public HSI datasets demonstrate that the proposed method accurately recognizes different objects with an overall accuracy (OA) of 86.68%, 94.73%, and 95.14% based on the pixel method, and 98.15%, 99.86%, and 99.98% based on the pixel-patch method.

Background: Three-dimensional gait analysis plays a crucial role in the rehabilitation assessment of post-stroke hemiplegic patients. However, the data generated from such analysis is often complex and difficult to interpret in clinical practice, requiring extensive time and complicated procedures. The Gait Deviation Index (GDI) serves as a simplified metric for quantifying the severity of pathological gait. Although isokinetic dynamometry is widely employed in muscle function assessment and rehabilitation, its application in gait analysis remains underexplored. Objective: This study aims to investigate the use of isokinetic muscle strength data obtained from the Biodex system, combined with machine learning techniques, to predict the GDI in hemiplegic patients. Methods: A cohort of 150 post-stroke hemiplegic patients was evaluated. The peak torque, peak torque/body weight, maximum work of repeated actions, coefficient of variation, average power, total work, acceleration time, deceleration time, range of motion, and average peak torque for both flexor and extensor muscles on the affected side were measured at three angular velocities (60°/s, 90°/s, and 120°/s) using the Biodex System 4 Pro. GDI was calculated using data from a Vicon three-dimensional motion capture system. The study employed four machine learning models—Lasso regression, random forest (RF), support vector regression (SVR), and BP neural networks—to model and validate the data. Model performance was evaluated using mean squared error (MSE), the coefficient of determination (R²), and mean absolute error (MAE). SHapley Additive exPlanations (SHAP) analysis was used to enhance model interpretability. Results: The RF model outperformed others in predicting GDI, with an MSE of 16.18, an R² of 0.89, and an MAE of 2.99. In contrast, the Lasso regression model yielded an MSE of 22.29, an R² of 0.85, and an MAE of 3.71. The SVR model had an MSE of 31.58, an R² of 0.82, and an MAE of 7.68, while the BP neural network exhibited the poorest performance with an MSE of 50.38, an R² of 0.79, and an MAE of 9.59. SHAP analysis identified the maximum work of repeated actions of the extensor muscles at 60°/s and 120°/s as the most critical features for predicting GDI, underscoring the importance of muscle strength metrics at varying speeds in rehabilitation assessments. Conclusion: This study highlights the potential of machine learning techniques in analyzing complex clinical data. The developed GDI prediction model, based on isokinetic dynamometry data, offers a novel, streamlined, and effective tool for assessing rehabilitation progress in post-stroke hemiplegic patients, with promising implications for broader clinical application.

Fluorescence microscopic images of cells contain a large number of morphological features that are used as an unbiased source of quantitative information about cell status, through which researchers can extract quantitative information about cells and study the biological phenomena of cells through statistical and analytical analysis. As an important research object of phenotypic analysis, images have a great influence on the research results. Saturation artifacts present in the image result in a loss of grayscale information that does not reveal the true value of fluorescence intensity. From the perspective of data post-processing, we propose a two-stage cell image recovery model based on generative adversarial network to solve the problem of phenotypic feature loss caused by saturation artifacts. The model is capable of restoring large areas of missing phenotypic features. In the experiment, we adopt the strategy of progressive restore to improve the robustness of the training effect, and add the contextual attention structure to enhance the stability of the restore effect. We hope to use deep learning methods to mitigate the effects of saturation artifacts to reveal how chemical, genetic, and environmental factors affect cell state, providing an effective tool for studying the field of biological variability and improving image quality in analysis.

With prosperous development and successful application of deep learning technologies in the field of remote sensing, numerous deep-learning based methods have emerged for synthetic aperture radar (SAR) automatic target recognition (ATR) tasks over the past few years. Generally, most deep-learning based methods can achieve outstanding recognition performance in condition that an abundant of labeled samples are available to train the model. However, in real application scenarios, it is difficult and costly to acquire and to annotate abundant SAR images due to the imaging mechanism of SAR, which poses a big challenge to existing SAR ATR methods. Thereby, SAR target recognition under the situation of few-shot, where only a scarcely few of labeled samples are available, is a fundamental problem that needs to be solved. In this paper, we put forward a new method named enhanced prototypical network with customized region-aware convolution (CRCEPN) to specially tackle the few-shot SAR ATR tasks. To be specific, We first develop a feature extraction network based on a customized and region-aware convolution, which can adaptively adjust convolutional kernels and their receptive fields according to each SAR image’s own characteristics as well as the semantical similarity among spatial regions, thus augmenting its capability to extract more informative and discriminative features. To achieve accurate and robust target identity prediction under the few-shot condition, we then propose an enhanced prototypical network which can improve the representation ability of the class prototype by properly making use of training and test samples together, thus effectively raising the classification accuracy. Meanwhile, a new hybrid loss is designed to learn a feature space with both inter-class separability and intra-class tightness as much as possible, which can further upgrade the recognition performance of the proposed method. Experiments performed on the moving and stationary target acquisition and recognition (MSTAR) dataset and the OpenSARShip dataset demonstrate that the proposed method is competitive with some state-of-the-art methods for few-shot SAR ATR tasks.

Named Entity Recognition (NER) is a fundamental and pivotal stage in the development of various knowledge-based support systems, including knowledge retrieval and question-answering systems. In the domain of pig diseases, Chinese NER models encounter several challenges such as the scarcity of annotated data, domain-specific vocabulary, diverse entity categories, and ambiguous entity boundaries. To address these challenges, we propose PDCNER, a Pig Disease Chinese Named Entity Recognition method leveraging lexicon-enhanced BERT and contrastive learning. Firstly, we construct a domain-specific lexicon and pre-train word embeddings in the pig disease domain. Secondly, we integrate lexicon information of pig diseases into the lower layers of BERT using a Lexicon Adapter layer, which employs char-word pair sequences. Thirdly, to enhance feature representation, we propose a lexicon-enhanced contrastive loss layer on top of BERT. Finally, a Conditional Random Field (CRF) layer is employed as the model's decoder. Experimental results show that our proposed model demonstrates superior performance over several mainstream models, achieving a precision of 87.76%, a recall of 86.97%, and an F1-score of 87.36%. The proposed model outperforms BERT-BiLSTM-CRF and LEBERT by 14.05% and 6.8%, respectively, with only 10% of the samples available, showcasing its robustness in data scarcity scenarios. Furthermore, the model exhibits generalizability across publicly available datasets. Our work provides reliable technical support for the information extraction of pig diseases in Chinese and can be easily extended to other domains, thereby facilitating seamless adaptation for named entity identification across diverse contexts. The codes have been open-sourced at https://github.com/tufeifei923/pdcner.

The working environment of mining electric locomotives is wet and muddy coal mine 1 roadway. There may be idling or slipping between the wheels and rails of mining electric locomotives 2 due to low friction between the wheel and rail and insufficient utilization of creep rate. Therefore, it 3 is necessary to control the creep rate within a reasonable range. In this paper, the autonomous control 4 algorithm for mining electric locomotives based on improved ε-greedy is theoretically proven to be 5 convergent and effective firstly. Secondly, after analyzing the contact state between the wheel and rail 6 under wet and slippery road conditions, it was concluded that the value of creep rate is an important 7 factor affecting the autonomous driving of mining electric locomotives. Therefore, the autonomous 8 control method for mining electric locomotives based on creep control is proposed in this paper. 9 Finally, the effectiveness of the proposed method was verified through simulation. The problem of 10 wheel slipping and idling caused by insufficient friction of mining electric locomotives in coal mining 11 environments is effectively suppressed. Autonomous operation of vehicles with optimal driving 12 efficiency can be achieved through quantitative control and utilization of the creep rate between 13 wheels and rails

The detection accuracy of bridge superficial defect using the deep neural network approach decreases significantly under light variation and weak texture condition. To address these issues, an enhanced intelligent detection method based on the YOLOv8 deep neural network is proposed in this study. Firstly, multi-branch coordinate attention (MBCA) is proposed to improve the accuracy of coordinate positioning by introducing a global perception module in coordinate attention mechanism. Furthermore, a deformable convolution based on MBCA is developed to improve the adaptability for complex feature shapes. Lastly, the deformable convolutional network-attention-YOLO (DAC-YOLO) detection algorithm is formed by replacing the deep C2F structure in the YOLOv8 architecture with deformable convolution. A supervised dataset consisting of 4794 bridge surface damage images is employed to verify the proposed method, and results show that it achieves improvements of 2.0% and 3.4% in mAP and R. Meanwhile, the model complexity decreases by 1.2G, increasing the detection speed by 3.5/f·s-1.

The objective of image super-resolution is to reconstruct a high-resolution (HR) image with the prior knowledge from one or several low-resolution (LR) images. However, in the real world, due to the limited complementary information, the performance of both single-frame and multi-frame super-resolution reconstruction degrades rapidly as the magnification increases. In this paper, we propose a novel two-step image super resolution method concatenating multi-frame super-resolution (MFSR) with single-frame super-resolution (SFSR), to progressively upsample images to the desired resolution. The proposed method consisting of an L0-norm constrained reconstruction scheme and an enhanced residual back-projection network, integrating the flexibility of the variational model-based method and the feature learning capacity of the deep learning-based method. To verify the effectiveness of the proposed algorithm, extensive experiments with both simulated and real world sequences were implemented. The experimental results show that the proposed method yields superior performance in both objective and perceptual quality measurements, compared to the baseline super-resolution algorithms in the cascade model. In addition, the experiment indicates that this cascade model can be robustly applied to different SFSR and MFSR methods.

Hyperspectral small target detection (HSTD) is a promising pixel-level detection task. However, due to the low contrast and imbalance number between the target and background spatially and high dimensions spectrally, it is a challenging one. To address these issues, this work proposes a representation learning-based graph and generative network for hyperspectral small target detection. The model builds a fusion network through frequency representation for HSTD, where the novel architecture incorporates irregular topological data and the spatial-spectral feature to improve its representation ability. Firstly, a graph convolutional network (GCN) module better models the non-local topological relationship between samples to represent the hyperspectral scene’s underlying data structure. The mini-batch-training pattern of the GCN decreases the high computational cost of building an adjacency matrix for high-dimensional data sets. In parallel, the generative model enhances the differentiation reconstruction and the deep feature representation ability with respect to the target spectral signature. Finally, a fusion module compensates for the extracted different types of HS features and integrates their complementary merits for hyperspectral data interpretation while increasing detection and background suppression capabilities. Experiments on different hyperspectral data sets demonstrate the advantages of the proposed architecture.

Colorectal cancer (CRC) with KRAS mutation is prone to develop metachronous metastasis. However, effective measures for prediction and prevention of such clinical issue are currently unavailable. In this study, we confirmed that KRAS mutation is a risk factor for CRC metachronous metastasis through meta-analysis and public database analysis. Based on the ICGC-AGRO database, we have constructed a risk scoring model for predicting metachronous metastasis in KRAS-mutant CRC using machine learning, which was validated in public datasets. Furthermore, we discovered that KRAS inhibitors can effectively suppress the migration and invasion capabilities of high risk CRC cells by wound healing assay and transwell experiment, which was validated ex vivo and in vivo using organoids and splenic liver metastatic mouse model. Finally, RNA sequencing, RT-qPCR, and western blots have proven that KRAS inhibitors can suppress the epithelial-mesenchymal transition (EMT) and transforming growth factor beta (TGF-beta) signaling pathway in CRC. Collectively, our study suggests that KRAS inhibitors may inhibit the metastatic ability of CRC by suppressing EMT and the TGF-beta pathway, suggesting their potential for the prevention of metachronous metastasis in CRC with KRAS mutation.

The study addresses the challenge of efficiently peeling pineapples, which have a distinct elliptical form, thick skin, and small eyes that are difficult to detect with conventional automated methods. This results in significant flesh waste. To improve the process, the researchers developed an integrated system combining an enhanced BlendMask method, termed SAAF-BlendMask, and a PCP method. SAAF-BlendMask improves the detection of small pineapple eyes, while PCP ensures accurate posture adjustment for precise path planning. The system uses 3D vision and deep learning technologies, achieving an average precision (AP) of 73.04% and a small object precision (APs) of 62.54% in eye detection, with a path planning success rate reaching 99%. The fully automated electromechanical system was tested on 110 real pineapples, demonstrating a reduction in flesh waste by 11.7% compared to traditional methods. This study highlights the potential of advanced machine vision and robotics in enhancing the efficiency and precision of food processing.

This research study focuses on addressing practical optimization challenges in audio-visual content using advanced computational intelligence methods, with a specific emphasis on privacy-preserving issues. The research will be presented as a proposal and can be conducted based on the provided ideas. Computational intelligence encompasses various approaches, including evolutionary algorithms, learning methods, fuzzy systems, and neural networks, which have become essential tools for tackling complex engineering optimization problems due to computational expenses, dimensional complexities, and interdisciplinary integration. The research objectives include a thorough investigation of optimization challenges in audio-visual content, the delineation of problem characteristics, and the development of relevant methodologies based on these specifications. The main goals of the study, titled "Advanced Computational Intelligence Assisted Less-Expensive Privacy-Preserving Optimization," encompass the development of a metamodel-based learning algorithm integrated with evolutionary computation for privacy-preserving optimization, less-expensive signal watermarking optimization using hybrid metamodeling and evolutionary algorithms, and a closed-loop robust intelligent PID control system applied in real-time dynamic signal watermarking. Additional objectives may be defined as the research progresses and further background review is conducted.

Chronic kidney disease (CKD) is a growing public health issue with significant morbidity and mortality rates. Traditional diagnostics, with about 70% accuracy, often delay CKD detection, highlighting the need for more efficient methods. Recent studies indicate a potential link between CKD and auditory health, yet this relationship remains underexplored due to methodological challenges and the complexity of establishing causality. Our research utilizes data spanning from 2000 to 2020 from the National Health and Nutrition Examination Survey (NHANES), covering 12,392 participants, including 2,060 diagnosed with CKD. Through meticulous analysis employing logistic regression and Mendelian randomization, we have unearthed novel insights into the bi-directional associations between hearing impairment and CKD. Furthermore, we developed and validated a machine learning model that surpasses traditional diagnostic approaches in terms of accuracy and predictive power. These findings highlight the innovative integration of auditory examinations with demographic data to enhance CKD detection. Our approach demonstrates the potential of machine learning in transforming diagnostic methodologies, thus offering a significant advancement in the field of nephrology and public health.

This study extends the CGAN-based MODIS cloud vertical-profile (64×64-scene, about 70km width×15km height) retrieval technique developed by Leinonen et al. (2019) [1] to construct seamless 3D cloud fields for the MODIS granules [2]. Firstly, the accuracy and spatial continuity of the Leinonen et al. [1] retrievals are statistically evaluated. Then, according to the characteristics of the retrieval error, a spatially overlapping-scene ensemble generation method and a bi-directional Ensemble Binning Probability Fusion (CGAN-BEBPF) technique are developed, which improves the CGAN retrieval accuracy and support to construct seamless 3D clouds. The CGAN-BEBPF technique involves three steps: cloud masking, intensity scaling, and optimal value selection. It ensures adequate coverage of the low reflectivity areas while preserving the high reflectivity cloud cores. The technique is applied to retrieve the 3D cloud fields of Typhoon Chaba and a multi-cell convective system. The cloud structures of the CGAN-BEBPF results are highly consistent with the ground-based radar observations. It can retrieve weak clouds at the top levels of convective storms and hurricanes that were missed by ground-based radars and fill the gaps of the ground-based radar lower-level PPI scan. Overall, CGAN-BEBPF can be automated to retrieve the rich informative 3D structures of typhoons and severe convective systems over the oceans along the MODIS swath, which certainly presents great value for improving maritime and near-shore typhoons and convection prediction and other cloud-sensitive applications.

Background: Clinical trails have revealed that patients with depression generally accompanied with learning and memory impairment, which critically impact on individual’s health and development. Danggui-Shaoyao-San (DSS), a famous Chinese complex prescription, significantly overcame depression and relieved cognitive disorder based on previous research and publication. However, its effectsand potential mechanism against chronic restraint stress (CRS) remained unknown. Methods: CRS animal model was established and mice were divided to six groups while they were oral administrated with Danggui-Shaoyao-San at doses of 1.25, 2.5 and 5 g/kg for 14 days. Emotional and cognitive performances were detected by behavior tests, meanwhile neural plasticity and its molecular mechanism were examined by Dil staining, western blot and immunofluorescence. Results: DSS treatment dose-dependently improved locomotion ability in open filed test, overcame depression behavior in forced swimming test and elevated plus maze test, enhanced learning and memory ability in Morris water maze test. CRS decreased number of total spines and mushroom spines, while DSS treatment dose-dependently restored these by Dil staining. Expression of BDNF and GluR1 were significantly down-regulated in CRS group, which were significantly normalized by DSS. Conclusions: DSS treatment dose-dependently reversed CRS-induced cognitive impairments by inducing structural remodeling of neurons.

In recent years, artificial intelligence technology has seen increasingly widespread application in the field of intelligent manufacturing, particularly with deep learning offering novel methods for recognizing geometric shapes with specific features. In traditional CNC machining, computer-aided manufacturing (CAM) typically generates G-code for specific machine tools based on existing models. However, the tool paths for most CNC machines consist of a series of collinear motion commands (G01), often resulting in discontinuities in the curvature of adjacent tool paths, leading to machining defects. To address these issues, this paper proposes a method for CNC system machining trajectory feature recognition and path optimization based on intelligent agents. This method employs intelligent agents to construct models, analyze the key geometric information in the G-code generated during CNC machining, and uses the MCRL deep learning model incorporating linear attention mechanisms and multiple neural networks for recognition and classification. Path optimization is then carried out using mean filtering, Bezier curve fitting, and an improved novel adaptive coati optimization algorithm (NACOA) according to the degree of unsmoothness of the path. Taking the optimization of a gear model’s process file as an example, the effectiveness of the proposed method is verified. The research results indicate that the CNC system machining trajectory feature recognition and path optimization method based on intelligent agents can significantly enhance the smoothness of CNC machining paths, reduce machining defects, and has substantial application value.

Federated learning is an effective approach for preserving data privacy and security, enabling machine learning to occur in a distributed environment and promoting its development. However, an urgent problem that needs to be addressed is how to encourage active client participation in federated learning. The Shapley Value, a classical concept in cooperative game theory, has been utilized for data valuation in machine learning services. Nevertheless, existing numerical evaluation schemes based on the Shapley Value are impractical as they necessitate additional model training, leading to increased communication overhead. Moreover, participants’ data may exhibit Non-IID characteristics, posing a significant challenge to evaluating participant contributions. Non-IID data weakens the marginal effects of participants, which in turn affects model accuracy and leads to underestimated contribution measurements. Current works often overlook the impact of heterogeneity on model aggregation. This paper presents a fair federated learning contribution measurement scheme that addresses the need for additional model computations. By introducing a noval aggregation weight, it enhances the accuracy of contribution measurement. Experiments on the MNIST dataset show that the proposed method can accurately compute the contributions of participants. Compared to existing baseline algorithms, the model accuracy is significantly improved, with a similar time cost.

As a combination of both concepts of innovation and environmental development, green innovation is of great significance to the sustainable development of the country and industry. Previous literatures have found the separate roles of green entrepreneurial orientation and inter-organizational learning in understanding green innovation issues. However, few studies have done a comprehensive analysis of integrating three streams of research: green entrepreneurial orientation, green innovation and supply chain learning capability. Based on the resource-based view and dynamic capability theory, we examine the direct of green entrepreneurial orientation on green innovation as well as indirect effect through the mediation of supply chain learning capability. Meanwhile, an empirical data set of 228 manufacturing companies in China (Shaanxi, Guangdong, Hebei, Jiangsu, and Shandong) was used to test our hypotheses. Findings from our empirical study suggest that supply chain learning capability partially mediates the positive relationships between green entrepreneurial orientation and its two consequences—green incremental innovation and green radical innovation. In addition, this research implies that when enterprise has a strong green entrepreneurial orientation, the enterprise should make an effort to enhance the level of supply chain learning capability so as to fully develop their green innovation.

Heavy rains have created significant threats to human health and life. Floods and other natural disasters, which have a global impact annually, can be attributed to extended durations of intense precipitation. Accurate rainfall predictions are crucial in nations such as Bangladesh, where agriculture is the predominant occupation. The efficiency of machine learning (ML) methods is enhanced by the non-linearity of rainfall, surpassing the effectiveness of other alternative ways. Machine learning techniques show that individual classifiers exhibit worse accuracy than ensemble learning (EL) methodologies. Ensemble Learning techniques are utilized for rainfall prediction and estimating rainfall quantity and daily average temperature to enhance comprehension of the diverse Machine Learning algorithms. This research implements the machine learning techniques and ensemble-based classifier to predict the rainfall occurrence, along with the machine learning regressor models and ensemble-based regressor for the rainfall amount prediction and daily average temperature prediction, using Bangladesh Weather Dataset. The results of the machine learning and ensemble-based models are compared using the Accuracy and F1 score for rainfall occurrence prediction. In contrast, MAE and RMSE evaluation metrics are used for ensemble regressor and regression algorithms for the rainfall amount and daily average temperature prediction. With an accuracy of 83.41% and a recall of 78.17%, the Ensemble Classifier is the best at predicting when it will rain, but its precision of 51.16% stands in for the lowest. The Ensemble Regression model outperforms Linear Regression, Random Forest, and SVR in rainfall amount prediction, with the lowest MAE of 0.36 and RMSE of 0.90. The Ensemble Regression provides the most precise results for daily average temperature prediction with the lowest MAE 0.42 and RMSE 0.54 highlighting its superiority over the other regression models in forecasting temperature with less error. Ensemble approaches consistently lead task performance metrics.

Early diagnosis of paroxysmal atrial fibrillation (PAF) could suggest patients to receive timely interventions in clinical practice. Various PAF onset prediction algorithms might benefit from accurate heart rate variability (HRV) analysis and machine learning classification but to be challenged in real-time monitoring scenarios. The aim of this study is to present an end-to-end deep learning based PAFNet model that integrates a sliding window technique on raw R-R interval of electrocardiogram (ECG) segments to achieve a real-time prediction of PAF onsets. This integration enable the deep convolutional neural network (CNN) to be customized as a light-weight architecture that accommodates the size of sliding windows simply by altering the input layer, specifically its effectiveness in making a new prediction with each new heartbeat. Catering to potential influence of input sizes, three CNN models were trained using 50, 100, and 200 R-R intervals, respectively. For each model, the performance of automated algorithms was evaluated for PAF prediction using a ten-fold cross-validation. As a results, a total of 56,381 PAFN-type and 56,900 N-type R-R interval segments was collected from publicly accessible ECG databases, and a promising prediction performance of the automated algorithm with a 100 R-R intervals was achieved in terms of the sensitivity of 97.12%, the specificity of 97.77%, and the accuracy of 97.45%, respectively. Importantly, the automated algorithm with a sliding windows step of 1 could process one sample in only 23.1 milliseconds and identify the onset of PAF at least 45 minutes in advance. The present results suggest that the sliding window technique on raw R-R interval sequences along by deep learning based algorithms may offers the possibility of providing an accurate, real-time, and end-to-end clinical tool for mass monitoring of PAF.

The next generation network requires not only ultra high data rate, global coverage and connectivity, but also reducing network deployment costs and energy consumption. The emergence of reconfigurable intelligent surface (RIS) technology provides an effective way to improve efficiency and reduce cost, while the passive elements bring new challenges of channel estimation (CE) and beam tracking. For a RIS-aided multiple-input and single-output (MISO) system, in this paper, to achieve the channel state information (CSI), we propose a principle component analysis (PCA) based staged channel estimation method. Based on the estimated channel, we propose a deep learning (DL) based beam tracking scheme to realize low-complexity RIS reflection coefficient design, which effectively improves the signal-to-noise ratio (SNR) at the user side. Simulation results verify our proposed channel estimation scheme based on PCA and beam tracking scheme based on DNN for the semi-active RIS-aided MISO systems can obtain approximate performances with much lower computational complexity.

The acquisition of vegetation coverage information is crucial for crop field management, and utilizing visible light spectrum vegetation indices to extract vegetation coverage information is a commonly used method. However, most visible light spectrum vegetation indices do not fully consider the relationships between the red, green, and blue bands during their construction, making it difficult to ensure accurate extraction of coverage information throughout the crop's entire growth cycle. To rapidly and accurately obtain potato vegetation coverage information, drones were used in this study to obtain high-resolution digital orthoimages of potato growth stages. Based on the differences in the grayscale values of potato plants, soil, shadows, and drip irrigation belts, this study presents a combination index of blue and green bands (BGCI) and a combination index of red and green bands (RGCI). The vegetation index intersection method was used with 10 vegetation information indices to extract vegetation coverage, and the differences in extraction accuracy were compared with those of the maximum entropy method and bimodal histogram method. Based on the high-precision fraction vegetation coverage (FVC) extraction results, the Pearson correlation coefficient method and random forest feature selection were used to screen 10 vegetation and 24 texture features, and the top 6 vegetation indices most strongly correlated with the FVC were selected for potato growth stage FVC estimation and accuracy verification. A high-precision potato vegetation coverage estimation model was successfully established. This study revealed that during the potato tuber formation and expansion stages, the BGCI combined with the vegetation index intersection method achieved the highest vegetation coverage extraction accuracy, with overall accuracies of 99.61% and 98.84%, respectively. The RGCI combined with the vegetation index intersection method achieved the highest accuracy, 98.63%, during the maturation stage. For the potato vegetation coverage estimation models, the model based on the BGCI achieved the highest estimation accuracy (R2=0.9116, RMSE=5.7903), and the RGCI also achieved good accuracy in terms of vegetation coverage estimation (R2=0.8987, RMSE=5.8633). In the generality verification of the models, the R2 values of the FVC estimation models based on the BGCI and RGCI were both greater than 0.94. A potato vegetation coverage estimation model was constructed based on two new vegetation information indices, demonstrating good accuracy and universality.

The joint extraction of entities and relations is a critical task in information extraction, and its performance directly affects the performance of downstream tasks. However, existing joint extraction models based on deep learning exhibit weak processing capabilities for the phenomenon of multiple pronunciations of one character and multiple characters of one pronunciation when processing Chinese texts, resulting in performance loss. To address these issues, this paper introduces part-of-speech (POS) and pinyin features to aid the model in learning semantic features that are more contextually appropriate. A Chinese Joint Entity and Relation Extraction Model based on Progressive Contrastive Learning and Heterogeneous Feature Fusion is proposed (CJE-PCHF). During model training, an interactive fusion network based on progressive contrastive learning is employed to learn the dependencies between pinyin, POS, and semantic features. This guides the model in heterogeneous feature fusion, capturing higher-order semantic associations between heterogeneous features. On the commonly used DuIE evaluation dataset for joint extraction, this model achieved a significant improvement, with the F1 score increasing by 5.4% compared to the benchmark model CasRel.

The IEEE 802.11ah standard is introduced to address the growing scale of Internet of Things (IoT) applications. To reduce contention and enhance energy efficiency within the system, the restricted access window (RAW) mechanism is introduced at the medium access control (MAC) layer to manage the significant number of stations accessing the network. However, the RAW parameters need to be appropriately determined to achieve optimal network performance. In this paper, we optimize the configuration of RAW parameters in the IEEE 802.11ah-based IoT network. We analyze and propose an RAW parameters optimization problem with the objective of improving network throughput and formulate it as a Markov decision process. To cope with the complex and dynamic network conditions, we propose a proximal policy optimization (PPO)-based deep reinforcement learning (DRL) algorithm to find the optimal RAW parameters. We construct network environments with periodic and random traffic in the NS-3 simulator to validate the performance of the proposed PPO-based RAW parameters optimization algorithm. Simulation results reveal that using the PPO-based DRL algorithm, optimal RAW parameters can be obtained under different network conditions, thereby significantly improving network throughput.

Generalized zero-shot learning (GZSL) aims to simultaneously recognize both seen classes and unseen classes by training only on seen class samples and auxiliary semantic descriptions. Recent state-of-the-art methods infer unseen classes based on semantic information or synthesize unseen classes using generative models based on semantic information, all of which rely on the correct alignment of visual-semantic features. However, they often overlook the inconsistency between original visual features and semantic attributes. Additionally, due to the existence of cross-modal dataset biases, the visual features extracted and synthesized by the model may also mismatch with some semantic features, which could hinder the model from properly aligning visual-semantic features. To address this issue, this paper proposes a GZSL framework that enhances the consistency of visual-semantic features using self-distillation and disentanglement network (SDDN). The aim is to utilize self-distillation and disentanglement network to obtain semantically consistent refined visual features and non-redundant semantic features to enhance the consistency of visual-semantic features. Firstly, SDDN utilizes self-distillation technology to refine the extracted and synthesized visual features of the model. Subsequently, the visual-semantic features are then disentangled and aligned using a disentanglement network to enhance the consistency of the visual-semantic features. Finally, the consistent visual-semantic features are fused to jointly train a GZSL classifier. Extensive experiments demonstrate that the proposed method achieves more competitive results on four challenging benchmark datasets (AWA2, CUB, FLO, and SUN).

An active power dispatch method for a microgrid (MG) with multi-type loads, renewable energy sources (RESs) and distributed energy storage devices (DESDs) is the focus of this paper. The MG operates in a grid-connected model, and distributed power sources contribute to the service for load demands. The outputs of multiple DESDs are controlled to optimize the active power dispatch. Our goal with optimization is to reduce the economic cost under time-of-use (TOU) price, and to adjust the excessively high or low load rate of distributed transformers (DTs) caused by the peak-valley demand and load uncertainties. To simulate a practical environment, the stochastic characteristics of multi-type loads are formulated. The transition matrix of system state is provided. Then, a finite-horizon Markov decision process (FHMDP) model is established to describe the dispatch optimization problem. A learning-based technique is adopted to search the optimal joint control policy of multiple DESDs. Finally, simulation experiments are performed to validate the effectiveness of the proposed method, and the fuzzification analysis of the method is presented.

A multimodal detection system with complementary capabilities for efficient detection was developed for impurity detection. The system consisted of a visible light camera, a multispectral camera, image correction and registration algorithms. It can obtain spectral feature and color feature at the same time, and has higher spatial resolution than a single spectral camera. This system was applied to detect impurities in Pu ’er tea to verify its high efficiency. The spectral and color features of each pixel in the images of Pu ’er tea were obtained by this system and used for pixel classification. The experimental results show that the accuracy of Support Vector Machine (SVM) model based on combined features is 93%, which is 7% higher than that based on only spectral features. By applying median filtering algorithm and contour detection algorithm to the label matrix extracted from pixel-classified images, 8 impurities except hair were detected successfully. Moreover, taking advantage of the high resolution of visible light camera, small impurities can be clearly imaged. By comparing the segmented color image with the pixel-classified image, small impurities such as hair could be detected successfully. Finally, it is proved that the system can obtain multiple images to allow a more detailed and comprehensive understanding of the detected items, and has excellent ability to detect small impurities.

Synthetic Aperture Radar (SAR) ship target detection has been extensively researched. However, most methods use the same dataset division for both training and validation. In practical applications, it is often necessary to quickly adapt to new loads, new modes, and new data to detect targets effectively. This presents a cross-domain detection problem that requires further study. This paper proposes a method for detecting SAR ships in complex backgrounds using fusion tensor and cross-domain adversarial learning. The method is designed to address the cross-domain detection problem of SAR ships with large differences between the training and test sets. Specifically, it can be used for the cross-domain detection task from the fully-polarised medium-resolution ship dataset (source domain) to the high-resolution single-polarised dataset (target domain). This method proposes a Channel Fusion Module (CFM) based on the YOLOV5s model. The CFM utilises the correlation between polarised channel images during training to enrich the feature information of single-polarised images extracted by the model during inference. The article proposes a module called Cross-Domain Adversarial Learning Module (CALM) to reduce overfitting and achieve adaptation between domains. Additionally, the paper introduces the Anti-Interference Head (AIH) which decouples the detection head to reduce the conflict of classification and localization problems. This improves the anti-interference and generalization ability in complex backgrounds. This paper conducts cross-domain experiments using the constructed medium-resolution SAR full polarization dataset (SFPD) as the source domain and the high-resolution single-polarised ship detection dataset (HRSID) as the target domain. Compared to the best-performing YOLOV8s model among typical mainstream models, this model improves Precision by 4.9%, Recall by 3.3%, AP by 2.4%, and F1 by 3.9%. This verifies the effectiveness of the method and provides a useful reference for improving cross-domain learning and model generalization capability in the field of target detection.

Chip pad alignment inspection is of great importance in the industrial field. However, due to the fact that chip pads are usually small, problems such as misdetection and missed detection often occur. When applying deep learning methods for chip pad detection, it is necessary to ensure accurate detection of small target chips while meeting the requirements of lightweight detection models for industrial needs. To solve the above problems, this paper proposes a lightweight model based on improved YOLOv5s. Firstly, the feature extraction part is improved to increase the network's focus on the target. Secondly, the feature fusion layer is improved to double the resolution of the prediction head, and the context-aware network is designed to enhance the context-capture ability of key features of small targets. Finally, SIoU is adopted as the loss function to improve the speed and accuracy of the regression frame. The experimental results show that the improved YOLOv5s algorithm improves the detection accuracy by 2.3% and reduces the network parameters by 81.8% compared to the original algorithm. The improved algorithm is combined with image processing techniques to design correction methods for alignment anomalies and realize real-time alignment anomaly correction in industry.

Classifying network flow subsequences in an Industrial IoT gateway is an effective method for identifying Industrial IoT faults. This paper proposes a subsequence classification algorithm SSGBUL-IKNN based on unsupervised learning encoding to address the problem of low classification accuracy. Firstly, we built a network module based on unsupervised learning to encode the flow sequence of the Industrial IoT gateway. In the encoding module, a subsequence calibration algorithm controls the increasing error during the coding process. Then, the specific fault type is obtained through an improved KNN classification algorithm based on the indication distance source from the integrated coding sequences converted by multi-dimensional network flow sequences. The test result on three Industrial IoT datasets of a sewage treatment plant shows that the accuracy of SSGBUL-IKNN in this paper exceeds 92%, significantly higher than the algorithm DTW and TSF. It reaches the classification requirements of Industrial IoT fault diagnosis.

The utilization of selective laser melting (SLM) for high entropy alloys (HEAs) holds significant promise in commercial applications, and substantial experimental research efforts have been directed toward this domain. To take advantage of the reported experimental data of SLM manufactured (SLM-ed) HEAs and reduce unnecessary experimentation, this study incorporates machine learning (ML) techniques for the phase and tensile properties prediction of SLM-ed HEAs, thus presenting a novel avenue for accelerating the discovery of new SLM-ed HEAs. Through the adjustment of material descriptors and machine learning models, a model has been developed with an impressive accuracy of 93.8% in distinguishing between face-centered cubic (FCC), body-centered cubic (BCC), and dual-phase (FCC+BCC). Additionally, optimized models have been devised for the prediction of tensile properties, namely ultimate tensile strength (UTS), yield strength (YS), and elongation (δ), achieving noteworthy MAPEs (mean absolute percentage errors) of 10.33%, 8.55%, and 28.48%, respectively. Furthermore, several HEAs were fabricated using SLM, and the experimental outcomes exhibited a favorable alignment with the predicted results. These efforts carry significant implications in advancing the utilization of SLM for HEAs.

Estimating the ocean’s subsurface thermohaline information from satellite measurements is essential for understanding ocean dynamics and El Niño phenomenon. This paper proposes an improved double-output Residual Neural Network (DO-ResNet) model to concurrently estimate the ocean subsurface temperature (OST) and ocean subsurface salinity (OSS) in the tropical Western Pacific using multi-source remote sensing data, including sea surface temperature (SST), sea surface salinity (SSS), sea surface height anomaly(SSHA), sea surface wind (SSW), and geographical information (including longitude (LON) and latitude (LAT)). In the model experiment, Argo data were used to train and validate the model, and the root mean square error (RMSE), normalized root mean square error (NRMSE) and coefficient of determination (R²) were employed to evaluate the model's performance. The results showed that the sea surface parameters selected in this study have a positive effect on the estimation process, and the average RMSE and R² values for estimating OST (OSS) by the proposed model are 0.34 °C (0.05 psu) and 0.91 (0.95), respectively. Under the data conditions considered in this study, DO-ResNet demonstrates superior performance relative to the extreme gradient boosting model, random forest model, and artificial neural network model. Additionally, this study evaluates the model’s accuracy by comparing its estimations of OST and OSS across different depths with Argo-derived data, demonstrating the model's ability to effectively capture the most spatial features, and by comparing NRMSE across different depths and seasons, the model displays strong seasonal adaptability. In conclusion, this research introduces a novel artificial intelligence technique for estimating OST and OSS in the tropical Western Pacific Ocean.

In order to solve the problem of fault diagnosis for CNC machine feed system under the condition of variable speed conditions, an intelligent fault diagnosis method based on multi-domain feature extraction and ensemble learning model is proposed. First, various monitoring signals including vibration signal, noise signal and current signal are collected. Then, the monitoring signals are preprocessed and the time-domain, frequency-domain and time-frequency domain feature indexes are extracted to construct a multi-dimensional mixed domain feature set. Finally, the feature set is putting into the constructed DoubleEnsemble-LightGBM model to realize the fault diagnosis of the feed system. The experimental results show that the model can achieve good diagnosis results under different working conditions for both the public data set and the feed system test bench data set, and the average overall accuracy is 91.07% and 98.06% respectively. Compared with XGBoost and other advanced ensemble learning models, the method has better accuracy. It provides technical support for the stable operation and intelligent of CNC machine tools..

Groundwater represents a pivotal asset in conserving natural water reservoirs for potable consumption, irrigation, and diverse industrial uses. Nevertheless, human activities intertwined with industry and agriculture contribute significantly to groundwater contamination, highlighting the critical necessity to appraise water quality for safe drinking and effective irrigation. This research primarily focused on employing the Water Quality Index (WQI) to gauge water's appropriateness for these purposes. However, the generation of an accurate WQI can prove time-intensive owing to potential errors in sub-index calculations. In response to this challenge, an artificial intelligence (AI) forecasting model was devised, aiming to streamline the process while mitigating errors. The study collected 422 data samples from Mirpurkash, a city nestled in the province of Sindh, for a comprehensive exploration of the region's WQI attributes. Furthermore, the study probed into unraveling the interdependencies amidst variables in the physiochemical analysis of water. Diverse machine learning classifiers were employed for WQI prediction, with findings revealing that random forest and gradient boosting eclipsed other algorithms, achieving an accuracy rate of 99%. In close pursuit were SVM and XGBoost, registering accuracy scores of approximately 95% and 93%, respectively, while KNN and Decision Trees garnered accuracy rates of 88% and 87%, respectively. In addition to WQI prediction, the study conducted an uncertainty analysis of the models using the R-factor, providing insights into the reliability and consistency of predictions. This dual approach, combining accurate WQI prediction with uncertainty assessment, contributes to a more comprehensive understanding of water quality in Mirpurkash and enhances the reliability of decision-making processes related to groundwater utilization

In the extensive monitoring of maritime traffic, maritime management frequently encounters incomplete automatic identification system (AIS) data, particularly in critical static fields like vessel types. This shortfall presents substantial challenges in safety management, necessitating robust methods for data completion. Distinct from road traffic, where road width imposes constraints, maritime vessels of identical classifications often choose significantly divergent routes, deviating by several nautical miles. Sole dependence on trajectory clustering is insufficient. Addressing this issue, we segments the maritime area into spatiotemporal grids, transforming vessel paths into sequences of grid encodings. Utilizing natural language processing techniques, we integrate the Word2vec word embedding approach with our novel biLSTM self-attention chunk-max pooling net (biSAMNet) model, enhancing the classification of vessel trajectories. This method is crucial for determining static details like vessel types. Employing the Taiwan Strait as a case study and benchmarking against CNN, RNN, and methods based on the attention mechanism, our findings underscore our model’s superiority. The biSAMNet achieves an impressive vessel classification F1 score of 0.94, using only 5-dimensional word embeddings, highlighting the effectiveness of Word2vec pre-trained embedding layers. This research introduces a novel paradigm for processing vessel trajectory data, greatly enhancing the accuracy of filling in incomplete vessel type details.

The solar energy can mitigate the power supply shortage in remote regions for portable irrigation systems. The accurate prediction of solar irradiance is crucial for determining the quantity of solar panels and the power capacity of photovoltaic power generation (PVPG) systems for mobile sprinkler machines. In this study, a prediction method is proposed to forecast the solar irradiance of typical irrigation areas. The impact of the meteorological parameters on the solar irradiance is studied, and four different parameter combinations are formed and considered as input to the prediction model. Based on meteorological data provided by ten typical radiation stations uniformly distributed nationwide, an Extreme Gradient Boosting (XGBoost) model optimized using the Whale Optimization Algorithm (WOA) is developed to predict the solar radiation. The prediction accuracy and stability of the proposed method are then evaluated for different input parameters through training and testing. The differences between the prediction performances of models trained based on single-station data and mixed data from multiple stations are also compared. The obtained results show that the proposed model achieves the highest prediction accuracy when the maximum temperature, minimum temperature, sunshine hours ratio, relative humidity, wind speed, and extraterrestrial radiation are used as the input parameters. The prediction effectiveness is also verified based on measured data. The WOA-XGBoost model has higher prediction accuracy than the XGBoost model, with predicted values falling within acceptable ranges. The model developed using mixed data of multiple stations can be applied for forecasting solar irradiance in different regions. This study provides a foundation for the optimization of the configuration of PVPG systems for mobile sprinkler machines.

Citizen science substantially contributes to large-scale biodiversity monitoring and the enhancement of public literacy. While both of the contributions largely depend on the public’s self- regulated learning. In this study, we employed a mix- method approach, incorporating both a questionnaire survey of 133 participants and interviews with 12 volunteers, aiming to explore the feature of self- regulated learning in citizen science, as well as identify the main factors impacting it. The volunteers are all different ages, from different jobs, with varying educational backgrounds. The findings revealed that the educational background of volunteers did not significantly impact their self-regulated learning. However, an increased frequency of participation was linked to higher levels of self-regulated learning. Furthermore, achievement goal orientation mediates the prediction of self-regulation by scientific epistemology. These insights underscore the importance of enhancing citizen science's impact on scientific literacy by fostering self-regulated learning among participants. This can be achieved through explicit instruction in scientific epistemology and by fostering a sense of achievement to boost achievement goal orientations. Additionally, our study advocates for the promotion of social interaction between students and out- of- school citizen scientists in school-based citizen science projects. The main innovation of this study is to consider volunteers as learners, not merely as contributors of data.

Artificial intelligence provides a new research concept for digital image processing. However, at present, artificial intelligence is rarely introduced into the teaching of digital image processing in colleges and universities, and there are problems such as obsolete teaching content, single teaching methods, and simple course experiments, which affect the teaching effect and are not conducive to the cultivation of comprehensive and innovative talents. Digital image processing technology brings more possibilities to communication engineering and makes communication more convenient. For example, video calls and photo transmission make people's communication methods in daily life more and more diversified. Time and space limitations allow people to meet online, creating more communication possibilities. However, there are still many problems and methods worthy of in-depth exploration. Therefore, this paper has a comprehensive understanding and mastery of the traditional and deep learning methods of digital image processing to improve the relevant project practice and scientific research exploration ability and refer for similar research conclusions.

Proteins need to be located in appropriate spatiotemporal contexts to carry out their diverse biological functions. Mislocalized proteins may lead to a broad range of diseases, such as cancer and Alzheimer’s disease. Knowing where a target protein resides within a cell will give insights into tailored drug design for a disease. As the gold validation standard, the conventional wet lab uses fluorescent microscopy imaging, immunoelectron microscopy, and fluorescent biomarker tags for protein subcellular location identification. However, the booming era of proteomics and high-throughput sequencing generates tons of newly discovered proteins, making protein subcel-lular localization by wet-lab experiments a mission impossible. To tackle this concern, in the past decades, artificial intelligence (AI) and machine learning (ML), especially deep learning methods, have made significant progress in this research area. In this article, we review the latest advances in AI-based method development in three typical types of approaches, including sequence-based, knowledge-based, and image-based methods. We also elaborately discuss existing challenges and future directions in AI-based method development in this research field.

While a raft of existing Chinese literature examines the associations between the outbreak of the pandemic and students’ mental health, rarely do Chinese studies assess the nuanced relationships between digital learning, parenting, and students’ mental health since the coronavirus (COVID-19) outbreak. Such a rarely discussed topic has substantial scholarly value as mismanagement of digital learning and parenting, such as the exposure to cyberbullying and negative parenting during the public health crisis, could add substantial, unforeseeable psychological burdens for Chinese students. In this article, the author applied a systematic review to find all relevant Chinese literature that contains the words “digital learning”, “children/adolescents”, “mental health”, and “parenting” published since January 2020. As such a complex topic has rarely been addressed in Chinese contexts, the author was only able to find four related scholarly articles. The author summarises the arguments and empirical findings to explore the nuanced relationships between a) digitalisation, isolation, parenting, and children’s mental health, b) parenting, teacher-student relationships, and students’ mental health, and c) maternal and paternal parenting styles.

Sparse training (ST) aims to ameliorate deep learning by replacing fully connected artificial neural networks (ANNs) with sparse or ultra-sparse ones, such as brain networks are, therefore it might benefit to borrow brain-inspired learning paradigms from complex network intelligence theory. Here, we launch the ultra-sparse advantage challenge, whose goal is to offer evidence on the extent to which ultra-sparse (around 1% connection retained) topologies can achieve any leaning advantage against fully connected. Epitopological learning is a field of network science and complex network intelligence that studies how to implement learning on complex networks by changing the shape of their connectivity structure (epitopological plasticity). One way to implement Epitopological (epi- means new) Learning is via link prediction: predicting the likelihood of non-observed links to appear in the network. Cannistraci-Hebb learning theory inspired the CH3-L3 network automata rule for link prediction which is effective for general-purpose link prediction. Here, starting from CH3-L3 we propose Epitopological Sparse Meta-deep Learning (ESML) to apply Epitopological Learning to sparse training. In empirical experiments, we find that ESML learns ANNs with ultra-sparse hyperbolic (epi-)topology in which emerges a community layer organization that is meta-deep (meaning that each layer also has an internal depth due to power-law node hierarchy). Furthermore, we discover that ESML can in many cases automatically sparse the neurons during training (arriving even to 30% neurons left in hidden layers), this process of node dynamic removal is called percolation. Starting from this network science evidence, we design Cannistraci-Hebb training (CHT), a 4-step training methodology that puts ESML at its heart. We conduct experiments on 7 datasets and 5 network structures comparing CHT to dynamic sparse training SOTA algorithms and the fully connected counterparts. The results indicate that, with a mere 1% of links retained during training, CHT surpasses fully connected networks on VGG16, GoogLeNet, ResNet50, and ResNet152. This key finding is an evidence for ultra-sparse advantage and signs a milestone in deep learning. CHT acts akin to a gradient-free oracle that adopts CH3-L3-based epitopological learning to guide the placement of new links in the ultra-sparse network topology to facilitate sparse-weight gradient learning, and this in turn reduces the convergence time of ultra-sparse training. Finally, CHT offers the first examples of parsimony dynamic sparse training because, in many datasets, it can retain network performance by percolating and significantly reducing the node network size. Our code is available at: https://github.com/biomedical-cybernetics/Cannistraci-Hebb-training.

Surface water plays a crucial role in climate change, human production, and life, making accurate monitoring and observation of surface water particularly important. However, due to the significant diversity and complexity of water distribution in surface space, accurate mapping of surface water faces considerable challenges. When extracting water bodies from medium-resolution satellite remote sensing images, CNN methods may suffer from limitations in receptive fields and insufficient context modeling capabilities, leading to the loss of water body boundary details and poor fusion of multiscale features. Currently, there is relatively little research on this issue; therefore, it is necessary to explore new combinations of deep learning networks to address these challenges. The purpose of this study is to address the above issues. We propose a new combination of deep learning networks that fully utilize multiscale information to enhance water features. Specifically, we first combine deformable convolutions with the Swin Transformer to increase effective receptive fields while better integrating global semantic information. This combination can capture features of water bodies at different scales, improve the accuracy and integrity of water extraction, and provide reliable technical support for detailed water body extraction. We tested the newly constructed model using Sentinel-2 satellite images. Our model achieved results of over 90%, with an average accuracy of 97.89%, average precision of 94.98%, average recall of 90.05%, and an average F1 score of 92.33%. In addition, our model achieved an accuracy of 98.03% in mountainous areas. Our experiments and results validate the potential of combining the Swin Transformer and deformable convolutions in detailed water body extraction.

Amidst the escalating demands for efficient Intelligent Transportation Systems (ITS) management, the significance of accurate Origin-Destination (OD) data prediction has emerged as a paramount concern, given its indispensable role in the ITS domain. This paper introduces a groundbreaking integrated model, creatively fusing the unique strengths of Graph Convolutional Network (GCN), Residual Neural Network (ResNet), and Long Short-Term Memory Network (LSTM) into a single architecture, which we name GraphResLSTM, to exploit road segment average speed data for OD prediction. Compared with conventional road segment traffic volume data, average speed data is more readily available, significantly reducing the data accessibility barrier for OD prediction. We underscore the critical role of selection of key road segments and employs a combination of Entropy Weight Method and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) to identify 41 representative road segments from the experimental road network. In the experimental phase, we simulate 86400000 seconds of traffic conditions, generating a total of 1440000 data records. Utilizing these simulated data, we design and conduct comparative experiments between models as well as data types. The results demonstrate that both the GraphResLSTM model and road segment average speed data outperform other models and data types notably in OD prediction tasks.

Based on the detailed exploration of AI techniques for the early identification of learning difficulties in education, this study underscores the pivotal role of educational data analysis in enhancing student outcomes. Educators can proactively identify and address learning challenges through systematic data collection and analysis, thereby minimizing long-term academic obstacles and emotional stress. The integration of AI, particularly through natural language processing and computer vision technologies, facilitates personalized learning paths and real-time interventions, offering tailored support that maximizes each student's learning potential. These advancements optimize educational resource allocation and foster a more inclusive and effective learning environment, paving the way for future advancements in personalized education and intelligent teaching practices.

Data security and user privacy concerns are increasingly gaining attention. Federated learning models based on differential privacy offer a distributed machine learning framework that protects data privacy; however, the added noise can impact the model's utility, making performance evaluation crucial. To optimize the balance between privacy protection and model performance, we propose the Adaptive Weight-Based Differential Privacy Federated Learning (AWDP-FL) framework. This framework processes model parameters from the perspectives of neural network layers and model weights. Initially, each participant trains the model locally. During iterative training, the clipping threshold is determined by selecting an adaptive weight coefficient, followed by adaptive gradient clipping to control the gradient magnitude. Subsequently, adaptive gradient updates are applied to the model, and dynamic Gaussian noise is introduced when uploading the parameters to protect participant privacy. The server aggregates these noise-perturbed parameters to update the global model. This framework ensures strong privacy protection while maintaining model accuracy. Theoretical analysis and experimental results validate the effectiveness of this framework under stringent privacy constraints.

Digital twin technology is revolutionizing traditional manufacturing paradigms. In modern manufacturing systems, digital twin technology is fraught with challenges due to the scarcity of labeled data. Specifically, existing supervised machine learning algorithms, with their reliance on voluminous training data, find their applicability constrained in real-world production settings. This paper introduces an unsupervised 3D reconstruction approach tailored for industrial applications, aimed at bridging the data void in creating digital twin models. Our proposed model, by ingesting high-resolution 2D images, autonomously reconstructs precise 3D digital twin models without the need for manual annotations or prior knowledge. Through comparisons with multiple baseline models, we demonstrate the superiority of our method in terms of accuracy, speed, and generalization capabilities. This research not only offers an efficient approach to industrial 3D reconstruction but also paves the way for the widespread adoption of digital twin technology in manufacturing.

Traditional classroom teaching model can no longer meet the needs of students' ability development. As a new model actively advocated in the modern education concept, personalized teaching focuses on the development needs of students and emphasizes respect for students' individualized growth. In this work, we utilize deep learning method to recommend teaching strategies for different features of students. Specifically, the proposed method includes the construction of a multi-layer neural network model, which can analyze students' multi-dimensional learning data and behavior patterns, so as to recommend the most suitable teaching strategies for individual students. Through data collection and preprocessing, model architecture design, model training and optimization, the model gradually improves the accuracy of prediction and recommendation. Experimental analysis verifies the effectiveness of the model through field teaching experiments, and the results show that the personalized teaching strategy based on deep learning can significantly improve the learning effect and participation of students.

Ship detection from synthetic aperture radar (SAR) imagery is crucial for various fields in real-world applications. Numerous deep learning-based detectors have been investigated for SAR ship detection, which require a substantial amount of labeled data for training. However, SAR data annotation is time-consuming and demands specialized expertise, resulting in that deep learning-based SAR ship detectors struggle due to a lack of annotations. With limited labeled data, semi-supervised learning is a popular approach for boosting detection performance by excavating valuable information from unlabeled data. In this paper, a semi-supervised SAR ship detection network is proposed, termed as Multi-Teacher Dempster-Shafer Evidence Fusion Net-work (MTDSEFN). The MTDSEFN is an enhanced framework based on the basic teacher-student skeleton frame, comprising two branches: the Teacher Group (TG) and the Agency Teacher (AT). The TG utilizes multiple teachers to generate pseudo-labels for different augmentation versions of unlabeled samples, which are then refined to obtain high-quality pseudo-labels by using Dempster-Shafer (D-S) fusion. The AT not only serves to deliver weights of its own teacher to the TG at the end of each epoch, but also updates its own weights after each iteration, enabling the model to effectively learn rich information from unlabeled data. The combination of TG and AT guarantees both reliable pseudo-label generation and learning comprehensive diversity information from numerous unlabeled samples. Extensive experiments were performed on two public SAR ship datasets, and the results demonstrated the effectiveness and superiority of the proposed approach.

Due to the significant price difference, expensive oils like olive oil are often blended with cheaper edible oils. This practice of adulteration in edible oils, aimed at increasing profits for producers, poses a major concern for consumers. Furthermore, adulteration in edible oils can lead to various health issues impacting consumer well-being. In order to meet the requirements of fast, non-destructive, universal, accurate and reliable quality testing for edible oil, the oblique-incidence reflectivity difference (OIRD) method combined with machine learning algorithms was introduced to detect a variety of edible oils. The prediction accuracy of Gradient Boosting, K-Nearest Neighbor, and Random Forest models exceeded 95%. Experimental results indicate that the OIRD method can serve as a powerful tool for detecting edible oils.

Grain yield prediction affects policy making in various aspects such as agricultural production planning, food security assurance, and adjustment of foreign trade. Accurately predicting grain yield is of great significance in ensuring global food security. This paper is based on the MODIS remote sensing image data products from 2010 to 2020, and adds band information such as vegetation index and temperature to form composite remote sensing data as a data set. Aiming at the lack of models for large-scale forecasting and the need for human intervention in traditional models, this paper proposes a grain production estimation model based on deep learning. First, image cropping and yield mapping techniques are used to process the data to generate training samples. Then the channel and spatial attention mechanism (Convolutional Block Attention Module, CBAM) is added for extracting spatial information in different remote sensing bands to improve the efficiency of the model. Long Short-Term Memory (LSTM) neural networks is also added to obtain feature information in the time dimension. Finally, a national-scale grain yield prediction model is constructed. The proposed model was tested on data from 2018 to 2020 showing an average R2 of 0.940 and an average RMSE of 80,020 tons, indicating that it can predict Chinese grain yield better. The model proposed in this paper extracts grain yield information directly from the composite remote sensing data, and solves the problem of small-scale research and imprecise yield prediction in an end-to-end manner.

A high-resolution catalog for the 2014 Ms6.5 Ludian aftershocks was constructed based on the deep learning phase-picking model (CERP) and the seismic-phase association technology (PALM). A specific training strategy, which combines the advantages of the conventional short-long window average energy ratio algorithm (STA/LTA) and AI algorithm, is employed to retrain the CERP model. The P & S phases were accurately detected and picked on continuous seismic waveforms by the retained AI model. Hypoinverse and HypoDD were utilized for precise location of 3286 events. Compared to the previous results, our new catalog exhibits superior performances in terms of location accuracy and the number of aftershock events, thereby enabling a more de-tailed depiction of the deep-seated tectonic features. According to the distribution of aftershocks, it can be inferred that, (1) the seismogenic fault of Ludian earthquake is the NW-trending Baogunao-Xiaohe Fault. (2) the Ludian aftershocks interconnected the discontinuous NW-trending Baogunao-Xiaohe Fault, and intersected with the Zhaotong-Ludian Fault. (3) It suggests that the NE-trending Zhaotong-Ludian Fault may have been intersected by the NW-trending Baogunao -Xiaohe Fault, indicating that the Baogunao-Xiaohe Fault is likely a rela-tively young Neogene fault.

The research of object classification and part segmentation is a hot topic in computer vision. A considerable number of studies have been carried out about deep learning on 3D point clouds. However,it is challenging to achieve effective feature learning due to sparsity of point clouds. Recently, a variety of Transformers have been adopted to improve point cloud processing and display great potential. Nevertheless, large numbers of Transformer layers tend to incur huge computational and memory costs. PointNet++ is one of the most influential neural architectures for point cloud understanding. Although the accuracy of PointNet++ has been largely surpassed by recent networks, this does not mean that PointNet++ has no potential.Thus, this paper offer two major contributions that significantly improve PointNet++ performance. Firstly, we introduce a novel contextual feature extraction (CFE) block that significantly enhances the feature extraction capabilities of PointNet++ networks. Secondly, to further enhance feature fusion, we seamlessly integrate a mutual learning (ML) block into the network architecture. By embedding these two innovative blocks within each layer of the network, we not only enrich the network's functionality but also impart it with greater robustness and adaptability. The specific experiments were conducted on the S3DIS (6-fold cross-validation) and Modelnet40 datasets with 86.5% and 92.7% accuracy, respectively, which proved that our model is comparable or even better than most existing methods for classification and segmentation tasks.

Near-surface humidity (Qa) is a key parameter that modulates oceanic evaporation and influences the global water cycle. Remote sensing observations act as feasible sources for long-term and large-scale Qa monitoring. However, existing satellite Qa retrieval models are subject to apparent uncertainties due to model errors and insufficient training data. Based on in situ observations collected over the China Seas over the last two decades, a deep learning approach named Ensemble Mean of Target deep neural networks (EMTnet) was proposed to improve the satellite Qa retrieval over the China Seas for the first time. The EMTnet model outperforms five representative existing models by nearly eliminating the mean bias and significantly reducing the root-mean-square error in satellite Qa retrieval. According to its target deep neural networks selection process, the EMTnet model can obtain more objective learning results when the observational data are divergent. The EMTnet model was subsequently applied to produce a 30-year monthly gridded Qa data over the China Seas. It indicates that the climbing rate of Qa over the China Seas under the background of global warming are probably underestimated by current products.

With the advancement of urbanization in China, effective control of pollutant emissions and air quality have become important goals in current environmental management. Nitrogen dioxide (NO2), as a precursor of tropospheric ozone and fine particulate matter, plays a significant role in atmospheric chemistry research and air pollution control. However, the uneven ground monitoring stations and low temporal resolution of polar-orbiting satellites set challenges for accurately assessing near-surface NO2. To address this issue, a spatiotemporal refined NO2 retrieval model was established for China using the geostationary satellite Himawari-8. The spatiotemporal characteristics of NO2 were analyzed and its contribution factors were explored. Firstly, seven Himawari-8 channels sensitive to NO2 were selected by using the forward feature selection based on information entropy. Subsequently, a 2DCNN-LSTM network model was constructed, incorporating the selected channels and meteorological variables as retrieval factors to estimate hourly NO2 in China from March 2018 to February 2020 (with a resolution of 0.05°, per hour). The performance evaluation demonstrated that the full-channel 2DCNN-LSTM model had good fitting capability and robustness (R2=0.74, RMSE=10.93), and further improvements were achieved after channel selection (R2=0.87, RMSE=6.84). The 10-fold cross-validation results indicated that the R2 between retrieval and measured values was above 0.85, the MAE was within 5.60, and the RMSE was within 7.90. R2 varied between 0.85 and 0.90, showing better validation at mid-day (R2=0.89) and in spring and fall transition seasons (R2 =0.88 and R2 =0.90). To investigate the cooperative effect of meteorological factors and other air pollutants on NO2, statistical methods (Beta coefficients) were used to test the factor interpretability. Meteorological factors as well as other pollutants were analyzed. From a statistical perspective, PM2.5, Boundary Layer Height, and O3 were found to have the largest impacts on near-surface NO2, with each standard deviation change in these factors leading to 0.28, 0.24, and 0.23 in standard deviations of near-surface NO2, respectively. Findings of the study contribute to a comprehensive understanding of the spatiotemporal distribution of NO2 and provide a scientific basis for formulating targeted air pollution policies.

Timely and accurate information on tree species is crucial for the sustainable management of natural resources, forest inventory, biodiversity detection, and carbon stock calculation. The advancement of remote sensing technology and artificial intelligence has facilitated the acquisition and analysis of remote sensing data, resulting in more precise and effective classification of tree species. Multimodal remote sensing data and deep learning seem to be the current tree species classification research mainstream, whether or not. The current review on the remote sensing data and deep learning tree species classification methods perspectives to analyze the unimodal and multimodal remote sensing data and classification methods in this realm is missing. To bridge the gap, we search for major trends in the remote sensing data and tree species classification methods, provide a detailed overview of classic deep learning-based methods for tree species classification, and discuss the limitations.

The aim of the present study was to predict the response of non-small cell lung cancer (NSCLC) patients to immune checkpoint inhibitors (ICIs) by leveraging computed tomography (CT) images using deep learning techniques. Retrospectively, 624 sequential CT images were gathered from 156 patients at Jiangsu Province Hospital, along with their clinical data. The dataset was subsequently partitioned into three groups: training (n=547), validation (n=64), and test (n=64). Moreover, an external validation cohort included 37 CT images from patients at Nanjing Pukou Peoples' Hospital, accompanied by comprehensive clinical data. An advanced Video Vision Transformer (ViViT) model incorporating global self-attention was utilized to analyse patients treated with ICIs and predict their response. The ViViT model's efficacy was evaluated using a confusion matrix and a receiver operating characteristic curve (ROC). Notably, the ViViT model demonstrated predictive prowess for ICIs response, yielding respective areas under the receiver operating characteristic curve (AUC) of 0.74 (95% CI: 0.69-0.78), 0.74 (95% CI: 0.61-0.86), 0.76 (95% CI: 0.62-0.88), and 0.69 (95% CI: 0.5-0.87) in the training, validation, test, and external validation cohorts. The present study illustrates how a deep learning model can provide a non-invasive means to predict clinical outcomes in NSCLC patients undergoing ICIs, potentially transforming personalized treatment approaches for individuals with NSCLC.

Underwater visual detection technology plays a pivotal role in fields such as marine exploration. With the increasing demand for underwater monitoring, the quest for efficient and reliable methods for underwater target recognition has become particularly significant. To address this requirement, this study developed an innovative underwater object detection architecture based on YOLOv8, named YOLOv8-MU, aimed at significantly enhancing detection accuracy.By integrating the LarK module proposed in UniRepLKNet to optimize the backbone network, YOLOv8-MU aims to achieve a larger receptive field without increasing the model’s depth. Further, this research introduces C2fSTR, an innovative method that combines Swin Transformer with the C2f module. Additionally, we have incorporated the SPPFCSPC_EMA module, which combines Cross-Stage Partial Fast Spatial Pyramid Pooling (SPPFCSPC) with attention mechanisms, significantly improving the detection accuracy and robustness of various biological targets. Moreover, by introducing a fusion block based on DAMO-YOLO into the neck of the model, we further enhanced the model’s capability in multi-scale feature extraction. Finally, the adoption of the MPDIoU loss function, designed around vertex distance, effectively tackles the challenges of localization accuracy and boundary clarity in underwater organism detection. Experimental results on the URPC2019 dataset demonstrate that the YOLOv8-MU model achieved an [email protected] of 78.4%, marking improvements of 5.6%, 1.1%, and 4.0% over YOLOv5s, YOLOv7, and YOLOv8n respectively, indicating the leading performance (SOTA) of this method. On the other hand, further evaluation on the URPC2020 dataset confirmed the generalization capability of the YOLOv8-MU architecture, with its [email protected] reaching 80.4%, surpassing various models including YOLOv5x and YOLOv8n, showcasing the wide applicability of our proposed improved model architecture.

With advancements in digital technologies such as 5G communications, big data, and cloud computing, the components of network operation systems have become increasingly complex, significantly complicating system monitoring and maintenance. Correspondingly, automated log anomaly detection has become a crucial means to ensure stable network operation and protect them from malicious attacks or failures. Conventional machine learning and deep learning methods assume consistent distributions between the training and testing data, adhering to a closed-set recognition paradigm. Nevertheless, in realistic scenarios, systems may encounter new anomalies that were not present in the training data, especially in log anomaly detection. Inspired by evidential learning, we propose a novel anomaly detector called LogEDL, which supervises training of the model through an evidential loss function. Unlike traditional loss functions, the evidential loss function not only focuses on correct classification, but also quantifies the uncertainty of predictions. This enhances the robustness and accuracy of the model in handling anomaly detection tasks, while achieving functionality similar to open-set recognition. To evaluate the proposed LogEDL method, we conduct extensive experiments on three datasets, i.e., HDFS, BGL, and Thunderbird, to detect anomalous log sequences. Experimental results demonstrate that our proposed LogEDL achieves state-of-the-art performance in anomaly detection.

The challenging terrain and deep ravines characteristic of mountainous regions often result in slower path planning and suboptimal flight paths for Unmanned Aerial Vehicles (UAVs) when traditional meta-heuristic optimization algorithms are employed. This study investigates the application of an Improved Enhanced Snake Optimizer (IESO) for three-dimensional path planning in a simulated rugged mountainous terrain. The initialization process in the enhanced snake optimizer is refined by integrating the Chebyshev chaotic map. Additionally, a non-monotonic factor is introduced to modulate the "temperature," and a boundary condition is incorporated into the dynamic opposition learning mechanism. These modifications collectively reduce the likelihood of population convergence to local optima during optimization. The feasibility of IESO is validated through time complexity and global convergence analyses. Comparative simulation experiments benchmarked IESO against five state-of-the-art optimization algorithms across test functions and path-planning simulated scenarios. The experimental results indicate that the enhanced algorithm achieves superior optimization precision, faster convergence speeds, and improved quality in the planned trajectories by 30%.

This research presents the Cross-Cultural Intelligent Language Learning System (CILS), a novel approach integrating Artificial Intelligence (AI) into language education to enhance cross-cultural communication. CILS utilizes advanced AI technologies to provide adaptive, personalized learning experiences that cater to the unique linguistic and cultural backgrounds of each learner. By dynamically adjusting content and methodology, CILS significantly improves linguistic proficiency and cultural understanding, essential for effective global interactions. The implementation of CILS in platforms such as Busuu and HelloTalk has demonstrated marked improvements in engagement and communication skills among learners. Empirical studies validate the system's effectiveness in real-world settings, showing enhanced learner performance and increased intercultural competence. Additionally, the Technology Acceptance Model (TAM) applied confirms that the usability and perceived usefulness of AI-driven systems strongly influence learner acceptance and sustained use. This study not only underscores the potential of AI in transforming language education but also highlights the critical role of cultural sensitivity in designing educational technologies.

The current study aims to apply and compare the performance of six machine learning algorithms, including three basic classifiers: random forest (RF), gradient boosting decision tree (GBDT), and extreme gradient boosting (XGB), as well as their hybrid classifiers, using the logistic regression (LR) method (RF+LR, GBDT+LR, and XGB+LR), in order to map the landslide susceptibility of Zhangjiajie City, Hunan Province, China. First, a landslide inventory map was created with 206 historical landslide points and 412 non-landslide points, which was randomly divided into two datasets for model training (80%) and model testing (20%). Second, 15 landslide conditioning factors (i.e., altitude, slope, aspect, plane curvature, profile curvature, relief, roughness, rainfall, topographic wetness index (TWI), normalized difference vegetative index (NDVI), distance to roads, distance to rivers, land use/land cover (LULC), soil texture, and lithology) were initially selected to establish a landslide factor database. Thereafter, the multicollinearity test and information gain ratio (IGR) technique were applied to rank the importance of the factors. Subsequently, we used a series of metrics (e.g., accuracy, precision, recall, f-measure, area under the ROC (receiver operating characteristic) curve (AUC), kappa index, mean absolute error (MAE), and root mean square error (RMSE)) to evaluate the accuracy and performance of the six models. Based on the AUC values derived from the models, the GBDT+LR model with the highest AUC value (0.8168) was identified as the most efficient model for mapping landslide susceptibility, followed by the XGB+LR, XGB, RF+LR, GBDT, and RF models, which achieved AUC values of 0.8124, 0.8118, 0.8060, 0.7927, and 0.7883, respectively. The results from this study suggest that the stacking ensemble machine learning method is promising for use in landslide susceptibility mapping in the Zhangjiajie area and is capable of targeting the areas prone to landslides.

This study discusses the application of deep learning technology in network intrusion detection systems (IDS) and focuses on a new model named CNN-Focal. First, through the review of traditional IDS technology, it analyzes its limitations in dealing with complex network traffic. Then, the design principle of the CNN-Focal model is described in detail, which uses threshold convolution and SoftMax multi-class classification technology to effectively improve abnormal traffic detection's accuracy and efficiency. The experimental results show that CNN-Focal performs well on the open data set, demonstrating the potential and advantages of its application in the natural network environment and providing a new perspective and method for further research of deep learning in the field of network security in the future.

The advance of the medical field in the 21st century has influenced the development of many innovations, including in medical education. PBL (problem-based learning) is one of the pedagogical approaches that has been studying and implementing worldwide to attract medical students’ attention with their learning style and learning environment. China is the most significant “receiver” country of foreign students in Asia, and clinical medicine is one of the most chosen majors for undergraduate. In China, PBL has been implementing in some schools, but mostly with the local students and only a few implementations to the international students. International medical students in China faced a variety of challenges during their studies. In this review paper, we give the solution to adapt to the challenges by focusing on the implementation the integrated organ/system curriculum/syllabus with the PBL method to the school system with the support of the school stakeholders and students’ role.

Traffic anomaly detection is crucial for urban management, yet current research is often confined to small-scale endeavors. This study collected nine months of real-time Wuhan traffic monitoring data from Amap. We proposed Traffic-ConvLSTM, a multi-scale spatial-temporal technique based on long and short-term memory networks (LSTM) and convolutional neural networks (CNN) to effectively achieve long-term anomaly detection at the city level. First, we convert the traffic flow data into image representation, which enables the capture of the spatial correlation between traffic flow and roads and the correlation between traffic flow and roads, as well as the surrounding environment. Second, The model utilizes convolution kernels of different sizes to extract spatial features at road-level, regional-level, and city-level scales, while incorporating temporal features of different time steps to capture hourly, daily, and weekly dynamics. Additionally, varying weights are assigned to the convolution kernels and temporal features of varying spatio-temporal scales to capture the heterogeneous strengths of spatio-temporal correlations within patterns of traffic anomalies. The proposed Traffic-ConvLSTM model exhibits improved performance over existing techniques in the task of identifying long-term and large-scale traffic anomaly occurrences. Furthermore, the analysis reveals significant traffic anomalies during holidays and urban sporting events. The diverse travel patterns observed in response to various activities offer insights for large-scale urban traffic anomaly management, providing recommendations for city-level traffic control strategies.

Background: Ovarian, cervical, and endometrial cancers stand as fatal health killers for women's mental and physical health, especially affecting female reproductive systems. Exploiting weakly supervised learning and explainable AI techniques are crucial for fast, accurate, and robust automatic marker detection, efficient prevention, and primary treatment of gynecological tumors. Methods: With respect to the PRIMSA methodology, weakly supervised learning schemes and deep learning-based schemes are investigated in the cross-subject fields of clinical diagnostic imaging and explainable AI technology. Related methods, opening research problems, and challenging subjects are explored in the screening and treatment of gynecological tumors and performing cancer image diagnosis in the latest study. Results: Keynote approaches combining ultrasound medicine, explainable AI, and clinical imaging technology are summarized. In combination with learning-based methods with medical imaging, prospective insights are put forward for refreshed concepts on treatments in the area of gynecological oncology, with a feasible range of applications for their corresponding AI techniques. Conclusions: Explainable AI and deep learning-based approaches are capable of performing accurate classification between benign and malignant tumors, yielding a pathway to care for women's health, improving the survival rate of patients, pacing with disease prediction, and matching the strategic goal of "Healthy China 2030".

Aesthetic ability is a powerful tool that can enhance student learning outcomes in arts education. This study not only determines the effect of aesthetic ability on student learning but also determines the effects of different factors on improvement in students' aesthetic abilities. The data used to investigate the mentioned relationships have been obtained from randomly selected 322 undergraduate students of arts education at different universities in Chengdu, Sichuan, China. The reliability of the data has been tested through Cronbach’s Alpha and regression analysis has been employed to determine the relationships among variables. Study findings revealed that teacher role, clarity of teaching goal, scientific plan, students' guidance, future creation, learning reflection, participation in exhibitions, improvement of art, appreciation of artwork, and innovative teaching methods influence student learning positively and significantly. Further, the effect of improvement in aesthetic ability on student learning has been found to be positive and significant. Based on the findings, some important implications have been discussed at the end of the study.

Unmanned Underwater Vehicles (UUVs) are pivotal in ocean exploration, research, and various industrial activities such as marine mining and offshore engineering. However, traditional methods of navigating these vehicles face significant challenges, mainly due to the inefficacy of electromagnetic waves in water, leading to signal loss. To address these limitations, researchers are increasingly turned learning, an artificial intelligence technique capable of learning from data-enhancing underwater navigation, mainly through visual Simultaneous Localization and Mapping (SLAM). This paper explores integrating deep learning methodologies and sensor technologies to revolutionize underwater navigation for UUVs. Proprioceptive sensors, along with exteroceptive sensors, are crucial in accurately measuring and comprehending the underwater environment. Additionally, the paper provides detailed insights into the processes of underwater SLAM, camera-based underwater positioning systems, sonar systems for underwater navigation, and the utilization of Lidar in underwater navigation. Furthermore, it delves into applying deep learning techniques in underwater SLAM, offering a comprehensive understanding of the innovative processes driving advancements in underwater vehicle navigation. By leveraging these advancements, this research aims to improve underwater navigation systems' precision, reliability, and adaptability, thereby unlocking new frontiers in ocean exploration and industrial applications for UUVs.

The aim of this study is (a) to develop, test, and employ a combined method of unsupervised machine learning to objectively assess the condition of sports facility in primary schools (PSSFC) and (b) examine the examine the geographical and typological association with PSSFC. Based on the Sixth National Sports Facility Census (NSFC), six PSSFC indicators (indoor and outdoor facility included) were selected as the measurements and decomposed by using the t-stochastic neighbor embedding (t-SNE). Thereafter, the Fuzzy C-mean (FCM) algorithm was used to cluster the same type of PSSFC with selecting the optimum numbers of evaluation level. Overall 845 primary schools in Shanghai, China were recruited and tested by this combined approach of unsupervised machine learning. In addition, the two-way analysis of covariance was used to examine the location and types of school associated with PSSFC variables in each level. The combined method was found to have acceptable reliability and good interpretability, differentiating PSSFC into five gradient levels. The characteristics of PSSFC differ by the location and school type of individual school. Our findings are conducive to the regionalized and personalized intervention and promotion on the children’s physical activity (PA) upon the practical situation of particular schools.

Recent advancements in deep learning have spurred the development of numerous novel semantic segmentation models for land cover mapping, showcasing exceptional performance in delineating precise boundaries and producing highly accurate land cover maps. However, to date, no systematic literature review has comprehensively examined semantic segmentation models in the context of land cover mapping. This paper addresses this gap by synthesizing recent advancements in semantic segmentation models for land cover mapping from 2017 to 2023, drawing insights on trends, data sources, model structures, and performance metrics based on a review of 106 extracted articles. Our analysis identifies top journals in the field, including MDPI Remote Sensing, IEEE Journal of Selected Topics in Earth Science, and IEEE Transactions on Geoscience and Remote Sensing, IEEE Geoscience and Remote Sensing Letters, ISPRS Journal Of Photogrammetry And Remote Sensing as the leading journals. We find that research predominantly focuses on land cover, urban areas, precision agriculture, environment, coastal areas, and forests. Geographically, 35.29% of the study areas are located in China, followed by USA (11.76%), France (5.88 %), Spain (4%) and others. Sentinel-2, Sentinel-1, and Landsat satellites emerge as the most commonly used data sources. Benchmark datasets such as ISPRS Vaihingen &amp; Potsdam, LandCover.ai, DeepGlobe, and GID datasets are frequently employed. Model architectures predominantly utilize encoder-decoder, and hybrid convolutional neural network-based structures because of their impressive performances, with limited adoption of transformer-based architectures due to its computational complexity issue, and slow convergence speed. Lastly, this paper highlights existing key research gaps in the field to guide future research directions.

This study employs the reflection symmetry decomposition (RSD) method to extract polarization scattering features from ground object images, aiming to determine the optimal data input scheme for deep learning networks in polarimetric synthetic aperture radar classification. Eight distinct polarizing feature combinations are designed, and the classification accuracy of various approaches is evaluated using the classic convolutional neural networks(CNNs) AlexNet and VGG16. The findings reveal that the commonly employed 6-parameter input scheme, favored by many researchers, lacks the comprehensive utilization of polarization information and warrants attention. Intriguingly, leveraging the complete 9-parameter input scheme based on the polarization coherence matrix results in improved classification accuracy. Furthermore, the input scheme incorporating all 21 parameters from the RSD and polarization coherence matrix notably enhances overall accuracy and the Kappa coefficient compared to the other 7 schemes. This comprehensive approach maximizes the utilization of polarization scattering information from ground objects, emerging as the most effective CNN input data scheme in this study. Additionally, the classification performance using the second and third component total power values (P2 and P3) from the RSD surpasses the approach utilizing surface scattering power value (PS) and secondary scattering power value (PD) from the same decomposition.

To enhance the accuracy and efficiency of address parsing in traffic accident handling, this study designs a deep address parsing model based on BERT, named ASSPM. This model effectively addresses the limitations of existing deep learning models in Chinese address parsing tasks by integrating address spatial semantics. First, a lexicon-enhancement method based on hybrid representation is employed, transforming the spatial semantics extracted by the Address Spatial Semantics Extractor (ASSE) into a symbolic semantic space that can be integrated with the feature vector space of the deep model. This achieves effective fusion of spatial semantics and the deep model at the model level. Second, a lexicon adapter enhanced with spatial semantics is utilized to integrate spatial semantics into the Transformer layers of the BERT model, achieving semantic fusion at the BERT level. Finally, comparative experiments on a custom address dataset and a public Chinese address dataset validate the proposed method’s superior performance over other baseline methods. This research provides robust support for the application of deep address parsing models integrating address spatial semantics in traffic accident handling, promising to improve the efficiency and accuracy of traffic accident management.

Tackling the intricacies of network dismantling in complex systems poses significant challenges. This task has relevance across various practical domains, yet traditional approaches focus primarily on singular metrics, such as the number of nodes in the Giant Connected Component (GCC) or average pairwise connectivity. In contrast, we propose a unique metric that concurrently targets nodes with the highest degree and reduces the GCC size. Given the NP-hard nature of optimizing this metric, we introduce MaxShot, an innovative end-to-end solution that leverages graph representation learning and reinforcement learning. Through comprehensive evaluations on both synthetic and real-world datasets, our method consistently outperforms leading benchmarks in accuracy and efficiency. These results highlight MaxShot’s potential as a superior approach to address the network dismantling problem effectively.

Network Function Virtualization (NFV) allows for the dynamic provisioning of Virtualized Network Functions, adapting services to the complex and real-time network environment for enhanced network performance. Despite the benefits of NFV, virtual network functions (VNFs) migration and the energy consumption pose significant challenges due to the dynamic nature of the physical network, especially in the edge-cloud continuum network. Currently, people tackle this challenge by predicting the network traffic and then proactively migrating the VNFs using the predicted values. There are two challenges here. First, when SFCs belong to different network service providers, the historical network traffic data is held by the service providers, and they are reluctant to share the historical data due to privacy concerns. Without the historical network traffic data of each SFC, the resource provider that owns the underlying network cannot effectively predict the network traffic; in addition, without the feedback of the migration results of the underlying physical network, the network traffic prediction will be difficult to match the needs of the physical network. In this manuscript, we solve the problem of network traffic prediction in the NFV-based edge-cloud continuum network for multiple network service providers. We aim to improve migration performance and reduce energy consumption while ensuring resource and latency requirements. To protect the privacy of each network service provider, we apply an FL algorithm. After that, to improve the migration performance, we tailor the loss function. In the loss function, factors such as the number of migrations, the number of migration failures, and the energy consumption of the NFV-based edge-cloud continuum network are added. Additionally, to obtain the results of the SFC migration, we develop a mathematical model for the problem of multiple network service providers in the NFV-based network. The effectiveness of our algorithm is evaluated through extensive simulations, and the results show a significant reduction in the number of migrated nodes and energy consumption while improving the SFC acceptance ratio compared to state-of-the-art schemes which only use the difference between predicted and actual flows to define the loss function.

In this study, we explored the application of deep learning techniques for credit card fraud detection, aiming to improve the performance and reliability of anomaly detection methods in financial transactions. We first utilized the Isolation Forest algorithm, achieving a detection accuracy of 26% for the top 1000 transactions. Subsequently, we experimented with the Autoencoder algorithm, an unsupervised deep neural network model, which enhanced the detection accuracy to 33.6% in the best case despite some fluctuations. The results demonstrate deep learning models' strong feature extraction capability and adaptability, highlighting their potential to surpass traditional methods. However, the high imbalance in the dataset, with only 0.17% of transactions being fraudulent, poses a significant challenge. This study underscores the necessity for further experimentation and optimization of network structures and hyperparameters to achieve more stable and efficient fraud detection. The findings provide valuable insights and reference points for future research in financial fraud detection using deep learning methodologies.

With the widespread application of artificial intelligence technology in various industries, users' attention to privacy and data security has increased significantly. Federated learning, as a new technology paradigm combining privacy-enhanced computing and artificial intelligence, resolves the contradiction between data security and open sharing. This paper presents the benefits of federated learning in terms of privacy, real-time processing, model robustness, compliance and cross-industry applications. At the same time, when combined with Edge AI technology, federated learning promotes the decentralisation of intelligent systems, improving data privacy protection and model accuracy. This paper also discusses the application cases of federated learning in the medical field, through local data processing and model training, effectively protecting user privacy, realizing medical data sharing and model optimization, and promoting the development of artificial intelligence.

With the advancement of technology, signal modulation types are becoming increasingly diverse and complex. The phenomenon of signal time-frequency overlap during transmission poses significant challenges for the classification and recognition of mixed signals, including poor recognition capabilities and low generality. This paper presents a recognition model for fine-grained analysis of mixed signal characteristics, proposing a Geometry Coordinate attention mechanism and introducing a low-rank bilinear pooling module to more effectively extract signal features for classification. The model employs a residual neural network as its backbone architecture and utilizes the Geometry Coordinate attention mechanism for time-frequency weighted analysis based on information geometry theory. This analysis targets multiple-scale features within the architecture, producing time-frequency weighted features of the signal. These weighted features are further analyzed through a low-rank bilinear pooling module, combined with the backbone features, to achieve fine-grained feature fusion. This results in a fused feature vector for mixed signal classification. Experiments were conducted on a simulated dataset comprising 39,600 mixed signal time-frequency plots. The model was benchmarked against a baseline using a residual neural network. The experimental outcomes demonstrated an improvement of 9% in the exact match ratio and 5% in the Hamming score. These results indicate that the proposed model significantly enhances the recognition capability and generalizability of mixed signal classification.

Prochlorococcus, a cyanobacteria genus of the smallest and most abundant oceanic phototrophs, encompasses ecotype strains adapted to high-light (HL) and low-light (LL) niches. To elucidate the adaptive evolution of this genus, we analyzed 40 Prochlorococcus marinus ORFeomes, including two cornerstone strains, MED4 and NATL1A. Employing deep learning with robust statistical methods, we detected new protein family distributions in the strains and identified key genes differentiating HL and LL strains. HL strains harbor genes (ABC-2 transporters) related to stress resistance, such as DNA repair and RNA processing, while LL strains exhibit unique chlorophyll adaptations (ion transport proteins, HEAT repeats). We report variable, depth-dependent endogenous viral elements in the 40 strains. We constructed the ORFeomes of MED4 and NATL1A covering 99 % of the annotated protein-coding sequences of the two species, totaling 3976 cloned, sequence-verified ORFs. These comparative genomics analyses, paired with MED4 and NATL1A ORFeomes, will facilitate future genotype-to-phenotype mappings and systems biology exploration of Prochlorococcus ecology.

Color change is the most obvious characteristic of tomato ripening stage and an important indicator of tomato ripening condition, which directly affects the commodity value of tomato. To visualize the color change of tomato fruit in mature stage, a gated recurrent unit network with an encoder-decoder structure that dynamically simulates tomato growth and development with time-dependent lines using tomato color and shape as real-time data was proposed in this paper. Firstly, the .json file was converted into a mask.png file, the tomato mask was extracted, and the tomato was separated from the complex background environment, and the tomato growth and development data set was successfully constructed. Then, a network of gated recurrent units with encoder-decoder structure was constructed to predict the future growth trend of tomato under different greenhouse temperatures. The experimental results showed that for the gated recurrent unit network of encoder-decoder structure proposed, when the hidden layer number was 1 and hidden layer number was 512, a high consistency and similarity between the model predicted image sequence and the actual growth and development image sequence would be realized, and the structural similarity index measure was 0.746. It was proved that when the average temperature was 24.93℃, the average soil temperature was 24.06℃, and the average light intensity was 11.26 Klux, the environment was the most suitable for tomato growth. The environmental data-driven tomato growth model was constructed to explore the growth status of tomato under different environmental conditions, thus to understand the growth status of tomato in time. The statement provided a theoretical foundation for determining the optimal greenhouse environmental conditions to achieve tomato maturity. And it offered recommendations for investigating the growth cycle of tomatoes, as well as technical assistance for standardized cultivation in solar greenhouses.

In recent years, due to the detrimental impact of traditional gas-liquid refrigerants on the environment, the pursuit of solid-state refrigeration technology based on elastocaloric effect, as a promising sustainable alternative, has been promoted. However, the traditional trial-and-error method is inefficient and difficult to meet the material application requirements. In this work, machine learning (ML) was employed to accelerate the development of NiTi-based shape memory alloy (SMA) with excellent elastocaloric effect. By means of active learning, we identified 9 novel NiTi-based SMAs in four iterations, the phase transformation-induced entropy change (ΔS) of which are as high as over 90 J/kg·K-1. Moreover, the model exhibits good interpretability, allowing us to understand the relevance between the target and features from the aspect of mechanism.

Accurate forecasting of crop yields holds paramount importance in guiding decision-making processes related to breeding efforts. This study focused on the application of multi-sensor data fusion and machine learning algorithms based on unmanned aerial vehicles (UAVs) in wheat yield prediction. Five machine learning (ML) algorithms namely random forest (RF), partial least squares (PLS), ridge regression (RR), K-Nearest Neighbor (KNN) and eXtreme Gradient Boosting Decision Tree (XGboost) were utilized for multi-sensor data fusion, and three ensemble methods including the second-level ensemble methods (stacking and feature-weighted) and the third-level ensemble method (simple average) for wheat yield prediction. The 270 wheat hybrids were used as planting materials under full and limited irrigation treatments. A cost-effective multi-sensor UAV platform, equipped with red–green–blue (RGB), multispectral (MS), and thermal infrared (TIR) sensors, was utilized to gather remote sensing data. The results revealed that the XGboost algorithm exhibited outstanding performance in multi-sensor data fusion, with the RGB+MS+Texture+TIR combination demonstrating the highest fusion performance (R2=0.660, RMSE= 0.754). Compared with the single ML model, the employment of three ensemble methods significantly enhanced the prediction accuracy of wheat yield. Notably, the third-layer simple average ensemble method demonstrated superior performance (R2 = 0.733, RMSE= 0.668 t ha-1). It significantly outperformed both the second-layer ensemble methods of Stacking (R2= 0.668, RMSE= 0.673 t ha-1) and feature-weighted (R2= 0.667, RMSE= 0.674 t ha-1), thereby exhibiting superior predictive capabilities. This finding demonstrated that the third-layer ensemble method not only augments the predictive ability of the model but also fine-tuned the accuracy of wheat yield prediction through the employment of simple average ensemble learning. Consequently, it offers a novel perspective for crop yield prediction and breeding selection.

Synthetic Aperture Radar has been extensively used in numerous fields and can gather a wealth of information about the area of interest. This large-scene data-intensive technology puts a high value on automatic target recognition (ATR) which can free the utilizers and boost the efficiency. Recent advances in artificial intelligence have made it possible to create a deep learning-based SAR ATR that can automatically identify target features from massive input data. In the last 6 years, intensive research has been conducted in this area, however, most papers in the current SAR ATR field used recurrent neural network (RNN) and convolutional neural network (CNN)-varied models to deepen the regime’s understanding of the SAR images. To equip SAR ATR with updated deep learning technology, this paper tries to apply a lightweight vision transformer (LViT)-based model to classify SAR images. The entire structure was verified by an open- accessed SAR data set and recognition results show that the final classification outcomes are robust and more accurate in comparison with referred traditional network structures without even using any convolutional layers.