Currently, skin cancer is the most commonly diagnosed form of cancer in humans and is one of the leading causes of death in patients with cancer. Biopsy methods are an invasive research method and are not always available for primary diagnosis. Imaging methods have low accuracy and depend on the experience of the dermatologist. Artificial intelligence technologies can match and surpass visual analysis methods in accuracy, but they have the risk of a false negative response when a malignant pigmented lesion can be recognized as benign. One possible way to improve accuracy and reduce the risk of false negatives is to analyze heterogeneous data, combine different preprocessing methods, and use modified loss functions to eliminate the negative impact of unbalanced dermatological data. The paper proposes a multimodal neural network system with a modified cross-entropy loss function that is sensitive to unbalanced heterogeneous dermatological data. The accuracy of recognition in 10 diagnostically significant categories for the proposed system was 85.19%. The novelty of the proposed system lies in the use of cross-entropy loss when training the modified function with the help of weight coefficients. The introduction of weighting factors has reduced the number of false negative forecasts, as well as improved accuracy by 1.02-4.03 percentage points compared to the original multimodal systems. The introduction of the proposed multimodal system as an auxiliary diagnostic tool can reduce the consumption of financial and labor resources involved in the medical industry, as well as increase the chance of early detection of skin cancer.

In recent years, the demand for coffee has increased tremendously. During the production process, green coffee beans are traditionally screened manually for defective beans before they are packed into coffee bean packages; however, this method is not only time-consuming but also increases the rate of human error due to fatigue. Therefore, this paper proposed a lightweight deep convolutional neural network (LDCNN) for the quality detection system of green coffee beans, which combined depthwise separable convolution (DSC), squeeze-and-excite block (SE block), skip block, and other frameworks. To avoid the influence of low parameters of the lightweight model caused by the model training process, rectified Adam (RA), lookahead (LA), and gradient centralization (GC) were included to improve efficiency; the model was also put into the embedded system. Finally, the local interpretable model-agnostic explanations (LIME) model was employed to explain the predictions of the model. The experimental results indicated that the accuracy rate of the model could reach up to 98.38% and the F1 score could be as high as 98.24% when detecting the quality of green coffee beans. Hence, it can obtain higher accuracy, lower computing time, and lower parameters. Moreover, the interpretable model verified that the lightweight model in this work is reliable, providing the basis for screening personnel to understand the judgment through its interpretability, thereby improving the classification and prediction of the model.

Election Algorithm (EA) is a powerful metaheuristics model motivated by phenomena of the socio-political mechanism of the presidential election conducted in many countries. EA is selected as a topic of discussion due to its capability and robustness to carry out complex problems in the random-2SAT logic program. This paper utilizes a hybridized EA assimilated with the Hopfield neural network (HNN) in carrying out random logic program (HNN-R2SATEA). The efficiency of the proposed method was compared with the existing traditional exhaustive search (HNN-R2SATES) model and the recently introduced HNN-R2SATICA model. From the result obtained, clearly proven that based on our proposed hybrid model outperformed other existing model based on the Global Minima Ratio (ZM), Mean Absolute Error (MAE), Bayesian Information Criterion (BIC) and Execution Time (ET). The expected outcome portrays that the EA algorithm outperformed the other two algorithms in doing random-kSAT logic program. The results proved the robustness, effectiveness, and compatibility of the HNN-R2SATEA model.

Polarimetric Synthetic Aperture Radar (PolSAR) images encompass valuable information that can facilitate extensive land cover interpretation and generate diverse output products. Extracting meaningful features from PolSAR data poses challenges distinct from those encountered in optical imagery. Deep Learning (DL) methods offer effective solutions for overcoming these challenges in PolSAR feature extraction. Convolutional Neural Networks (CNNs) play a crucial role in capturing PolSAR image characteristics by leveraging kernel capabilities to consider local information and the complex-valued nature of PolSAR data. In this study, a novel three-branch fusion of complex-valued CNN named Shallow to Deep Feature Fusion Network (SDF2Net) is proposed for PolSAR image classification. To validate the performance of the proposed method, classification results are compared against multiple state-of-the-art approaches using the Airborne Synthetic Aperture Radar (AIRSAR) datasets of Flevoland, San Francisco, and ESAR Oberpfaffenhofen dataset. The results indicate that the proposed approach demonstrates improvements in OA, with 1.3% and 0.8% enhancement for the AIRSAR datasets and 0.5% improvement for the ESAR dataset. The analyses conducted on the Flevoland data underscore the effectiveness of the SDF2Net model, revealing a promising OA of 96.01% even with only 1% sampling ratio.

Brain-to-brain communication (B2B-C) is rapidly expanding, integrating communication technology and neuroscience to enable direct neural data transfer between people. Nonetheless, because neural data is susceptible to noise, interference, and hostile attacks, guaranteeing the security and resilience of B2B-C systems continues to be difficult. This work aims to use Adversarial Neural Network Training (ANNT) to improve the security of B2B-C systems by using Steady-State Visually Evoked Potentials (SSVEP) EEG data. We use two large SSVEP datasets for a thorough analysis: Lee2019_SSVEP and Nakanishi2015. We use the Fast Gradient Sign Method (FGSM) to create adversarial instances and ANNT to train the model on clean and adversarially perturbed data. The system's accuracy and resilience are significantly improved by ANNT, as seen by the up to 17% increase in adversarial accuracy and the average 0.03 point improvement in the Area Under the Curve (AUC). This work demonstrates how ANNT may strengthen B2B-C systems against advanced cyberattacks, opening the door for dependable and safe neural communication technologies.

The neural network technology for real-time cryptographic data protection with symmetric keys (masking codes, neural network architecture and weights matrix) for unmanned aerial vehicles (UAV) onboard communication systems has been developed. It provides hardware and software implementation with high technical and operational characteristics. The development of neural network technology for real-time cryptographic data protection was performed using an integrated approach based on theoretical foundations of neural network cryptographic data protection, new algorithms and structures of neural network data encryption and decryption, modern element base with the possibility of programming for the structure, and computer-aided design of hardware and software tools. The development and implementation of the on-board system of neural network cryptographic data protection in real-time are based on the following principles: variable composition of equipment; modularity; conveyorization and spatial parallelism; specialization and adaptation of hardware and software to data encryption and decryption. The tabular-algorithmic method of calculating the scalar product has been improved, it provides fast calculation of the scalar product of input data for both fixed and floating-point by bringing to the largest common order of weights and forming tables of macro-partial products for them. Components of neural network cryptographic data encryption and decryption have been developed on the processor core supplemented by specialized scalar product calculation modules. The specialized hardware for neural network cryptographic data encryption was developed using VHDL for equipment programming in the Quartus II development environment ver. 13.1 and the appropriate libraries, and implemented on the basis of the FPGA EP3C16F484C6 Cyclone III family.

Structure–function relationships in proteins have been one of the crucial scientific topics. Heme proteins have diverse and pivotal biological functions. Therefore, clarifying their structure–function correlation is significant to understand their functional mechanism and is informative for various fields of science. In this study, we constructed convolutional neural network models for predicting protein functions from the tertiary structures of heme-binding sites (active sites) of heme proteins to examine the structure–function correlation. As a result, we succeeded in the classification of oxygen-binding protein (OB), oxidoreductase (OR), proteins with both functions (OB–OR), and electron transport protein (ET) with high accuracy. Although the misclassification rate for OR and ET was high, the rates between OB and ET and between OB and OR were almost zero, indicating that the prediction model works well between protein groups with very different functions. However, predicting the function of proteins modified with amino acid mutation(s) remains a challenge. Our findings indicate a structure–function correlation in the active site of heme proteins. This study is expected to be applied to the prediction of more detailed protein functions such as catalytic reactions.

Great Malang region is developing rapidly with the population increase and inhabitant`s activity, like migration and urbanization. Other activities like agricultural expansion as well as an uncontrolled residential development need to be monitored to avoid any negative impact in the future. The availability of free and open-source software, spatial high-resolution satellite imagery datasets, and powerful algorithms open the possibilities to map, monitor, and predict the future trend of land use land cover (LULC) changes. However, the accuracy and precision of this model is still in doubt, especially in the Great Malang region. Research is needed to provide a foundational basis and documentation on how the changes occur, where did the changes occur, and the accuracy of the predicted model. This study tries to answer those questions using the high spatial resolution of Sentinel-2 imageries. Combination of the fuzzy algorithm, artificial neural network, and cellular automata was utilized to process the datasets. We analysed four different scenarios of simulation and the result then compared. The different number of hidden layers and iteration was used and evaluated to understand the effect of different parameters in the prediction result. The best scenario was then used to predict future land use changes. This study has successfully produced the future LULC model of Great Malang region with high accuracy level (87%). The study also found that the land use transformation from agriculture to urban built-up area is relatively low, where changes of the built-up area over three periods of analysis are below than 5%. This is due to the physical condition of Great Malang region where mountainous areas are dominated.

Large language models (LLMs) have shown impressive performance and versatility. However, their billions of parameters and high computational costs hinder the development of personalized and privacy-preserving AI assistants operating locally on user devices. In this work, we explored the potential of pruning LLMs to create lightweight models suitable for user devices, using the moderate-sized Llama-2 7B model as an example. By adopting a simple yet effective pruning method, we found that up to 60% of the weights in the Llama-2 7B model could be pruned without significantly impairing its language modeling capabilities. Furthermore, despite occasional factual inaccuracies, the pruned model at the sparsity limit generated fluent and helpful answers to daily queries, demonstrating its feasibility of on-device AI assistants. These inaccuracies might originate from forgetting or hallucination due to pruning. We proposed a simple protocol to distinguish between the two mechanisms, as well as future directions to improve the pruned models for local AI assistants.

Ultrasound (US) is commonly used for the diagnosis of liver masses. Ensemble learning has been widely used for image classification, but its methods have not been fully optimized. This study was performed to investigate the usefulness of ensemble learning and compare a number of ensemble learning techniques using multiple convolutional neural network (CNN)-trained models for image classification of liver masses in US images. The US imaging data set was classified into four categories: benign liver tumor (BLT, 6320 images), liver cyst (LCY, 2320 images), metastatic liver cancer (MLC, 9720 images), and primary liver cancer (PLC, 7840 images). In this study, 250 test images were randomly selected for each class, for a total of 1000 images, and the remaining images were used for training. Sixteen different CNNs were used for to train and test the US images. All four types of ensemble learning—soft voting (SV), stacking (ST), weighted average voting (WAV), and weighted hard voting (WHV)—showed greater accuracy than the single CNN. All four types also showed significantly better deep learning performance than ResNeXt101 alone. For image classification of liver masses using US images, ensemble learning improved the performance of deep learning over a single CNN.

Considering the significant advancements in autonomous vehicle technology, research in this field is of interest to researchers. To drive vehicles autonomously, controlling steer angle, gas hatch, and brakes need to be learned. The behavioral cloning method is used to imitate humans’ driving behavior. We created a dataset of driving in different routes and conditions and using the designed model, the output used for controlling the vehicle is obtained. In this paper, the Learning of Self-driving Vehicles Based on Real Driving Behavior Using Deep Neural Network Techniques (LSV-DNN) is proposed. We designed a convolutional network which uses the real driving data obtained through the vehicle’s camera and computer. The response of the driver is during driving is recorded in different situations and by converting the real driver’s driving video to images and transferring the data to an excel file, obstacle detection is carried out with the best accuracy and speed using the Yolo algorithm version 3. This way, the network learns the response of the driver to obstacles in different locations and the network is trained with the Yolo algorithm version 3 and the output of obstacle detection. Then, it outputs the steer angle and amount of brake, gas, and vehicle acceleration. The LSV-DNN is evaluated here via extensive simulations carried out in Python and TensorFlow environment. We evaluated the network error using the loss function. By comparing other methods which were conducted on the simulator’s data, we obtained good performance results for the designed network on the data from KITTI benchmark, the data collected using a private vehicle, and the data we collected.

Objective: The aim of this study was to provide a new machine learning method to determine temporal events and inner-cycle parameters (e.g., cycle, poles and skis contact and swing time) in cross-country roller ski skating on the field, using a single deported inertial measurement unit (IMU). Methods: The developed method is based on long short-term memory neural networks to detect poles and skis initial and final contact with the ground during the cyclic movements. Eleven athletes skied four laps of 2.5 km at low and high intensity using skis with two different rolling coefficients. They were equipped with IMUs attached to the upper back, lower back and to the sternum. Data from force insoles and force poles were used as reference system. Results: The IMU placed on the upper back provided the best results, as the LSTM network was able to determine the temporal events with an accuracy ranging from 49 to 55 ms and the corresponding inner-cycles parameters were calculated with a precision of 63 to 68 ms. The method detected 95% of the events for the poles and 87% of the events for the skis. Conclusion: The proposed LSTM method provides a promising tool for assessing temporal events and inner-cycle phases in roller ski skating showing the potential of using a deported IMU to estimate different spatio-temporal parameters of human locomotion.

Accurate segmentation of skin lesions from dermoscopic images is crucial for early skin cancer detection, yet variations in lesion appearance and image artifacts present challenges. This study proposes two deep learning models, UCM-NetV2 and BNN-UCM-NetV2, to improve accuracy and computational efficiency. UCM-NetV2 enhances the UCM-Net architecture with a novel "cyber-structure" combining Multilayer Perceptron and CNN layers, improving prediction accu-racy without increasing the model's parameter count. BNN-UCM-NetV2 further optimizes for mobile deployment by binarizing activations and weights into 1-bit values, reducing computa-tional complexity and model size. Evaluations on the ISIC2017 and ISIC2018 datasets demonstrate that UCM-NetV2 outperforms existing methods in both accuracy and computational efficiency, achieving this with less than 0.04 GFLOPs. These models have the potential to make skin lesion analysis accessible to a broader population, particularly in resource-limited settings, allowing for proactive skin health monitoring and facilitating teledermatology. To foster further innovation in mobile health diagnostics, the source code for UCM-NetV2 will be available on GitHub: https://github.com/chunyuyuan/UCMV2-Net.

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.

Recommender Systems (RecSys) are crucial in managing information overload and enhancing user satisfaction across various digital platforms, including e-commerce and entertainment. Evolving from traditional models to Deep Neural Networks (DNNs) and, more recently, Large Language Models (LLMs), these systems leverage sophisticated algorithms to analyze user behaviors and preferences. LLMs, such as GPT-4, are trained on extensive datasets to comprehend and generate natural language, significantly advancing their ability to deliver personalized recommendations. This tutorial explores the transformative impact of LLMs on RecSys, discussing their development, application in handling complex datasets, and the integration of contextual insights. Real-world examples illustrate how LLMs enhance recommendation accuracy and user experience, highlighting challenges and future directions in the field.

This paper represents a groundbreaking paradigm shift in network optimization. Departing from traditional static methodologies, this innovative approach harnesses the power of Generative Artificial Intelligence (AI) and Large Language Models (LLMs) to optimize cloud networks dynamically. By integrating advanced AI algorithms, this framework continuously adapts and evolves, ensuring optimal real-time performance. This dynamic optimization enhances efficiency and resilience, allowing cloud networks to adjust seamlessly to changing demands and conditions. Through the fusion of cutting-edge technology and adaptive intelligence, this approach heralds a new era in network optimization, empowering organizations to achieve unprecedent-ed levels of agility and scalability in their cloud infrastructures.

Variations in groundwater levels that are noticed in different parts of the world are typically the primary focus of research carried out in hydrology. The fluctuating levels of groundwater can be ascribed to several different factors, some of which include an increase in the demand for water, inappropriate irrigation practices, improper management of soil, and unregulated extraction from aquifers. It is necessary to effectively manage groundwater resources to have a trustworthy method of measuring and forecasting groundwater levels. The dynamics of groundwater are inherently uneven and difficult to understand. As a result, applying methods driven by data could potentially yield significant gains in hydrology. In this study, two data-driven models were utilized to estimate groundwater levels at a total of 4 monitoring wells located in Colorado state, USA. These models included ARIMA and ANN. The comparative analysis of the algorithms made use of a data set that had one month's worth of information from each of the years 1980 to 2019. The models were put through their paces, and the results of those tests were statistically and visually assessed. To evaluate the relative accuracy and precision of the models, the mean absolute error (MAE), mean square error (MSE), and root-mean-square error (RMSE) were used. When compared to ARIMA, the analysis demonstrates that ANN can produce the most accurate forecasts of groundwater levels in Colorado state USA.

For Convolutional Neural Network (CNN), Depthwise Separable CNN (DSCNN) is a preferred architecture for Application Specific Integrated Circuit (ASIC) implementation on edge devices. It can benefit from a multi-mode approximate multiplier proposed in this work. The proposed approximate multiplier uses two 4-bit multiplication operations to implement a 12-bit multiplication operation by reusing the same multiplier array. With this approximate multiplier, sequential multiplication operations are pipelined in a modified DSCNN to fully utilize the PE array in the convolutional layer. This Approximate (A-DSCNN) was implemented on TSMC 40-nm CMOS process with a supply voltage of 0.9 V. At the clock frequency of 200 MHz, the design achieves 4.78 GOPs/mW while occupying 1.24 mm x 1.24 mm silicon area. Compared to conventional DSCNN implemented in a similar process node, the chip area and power consumption were reduced by 53% and 25%, while the throughput was improved by 17%.

This study is based on the BP neural network regression model optimised by genetic algorithm and dedicated to predicting the amount of car purchase, this study is important for all parties in the automotive industry. By plotting the violins of six indicators, including gender, age, annual salary, credit card debt, net worth, and car purchase amount, we found that, except for gender, the other data showed a trend of dense in the middle gradually decreasing to the upper and lower sides, with an overall normal distribution. The correlation analysis shows that the amount of car purchase has a high positive correlation with age (0.63), followed by annual income (0.62), the correlation between net worth and the amount of car purchase is 0.49, while the correlation between credit card debt and the amount of car purchase is low (0.03). In the experimental part, we analyse the predicted-actual values of the training and test sets in comparison and find that the model predicted values are very close to the actual values. The RMSE of the training set is 11.559 and the RMSE of the test set is 11.4835, which is not much different from each other, indicating that the model has good generalisation ability. In addition, the R2 of both the training and test sets is 1, indicating that the model fits well. The MAE of the training set is 8.2512 and the MAE of the test set is 9.578; the MBE of the training set is -4.3035 and the MBE of the test set is -2.4547.These metrics indicate that the model performs well in predicting the amount of car purchase. In conclusion, this study provides an effective and reliable method to predict the amount of car purchase, creating a useful bridge between car manufacturers, dealers and consumers.

The passenger prediction flow is very significant to transportation sustainability. This is due to some chaos of traffic jams encountered by the road users during their movement to the offices, schools, or markets at earlier of the days and during closing periods. This problem is peculiar to the transportation system of the Federal University of Technology Minna, Nigeria. However, the prevailing technique of passenger flow estimation is non-parametric which depends on the fixed planning and is easily affected by noise. In this research, we proposed the development of a hybrid intelligent passenger frequency prediction model using the Auto-Regressive Integrated Moving Average (ARIMA) linear model, Convolutional Neural Network (CNN), and Kalman Filter Algorithm (KFA). The passengers’ frequency of arrival at the bus terminals is obtained and enumerated through the closed-circuit television (CCTV) and demonstrated using the Markovian Queueing Systems Model (MQSM). The ARIMA model was used for learning and prediction and compared the result with the combined techniques of using CNN-KFA. The autocorrelation coefficient functions (ACF) and partial autocorrelation coefficient functions (PACF) are used to examine the stationary data with different features. The performance of the models was analyzed and evaluated in describing the short-term passenger flow frequency at each terminal using the Mean Absolute Percentage Error (MAPE) and Mean Squared Error (MSE) values. The CNN-Kalman-filter model was fitted into the short-term series and the MAPE values are below 10%. The Mean Square Error (MSE) shows that the CNN-Kalman Filter model has the overall best performance with 83.33% of the time better than the ARIMA model and provides high accuracy in forecasting.

Functional connectivity (FC) is a potential candidate that can increase the performance of brain-computer interfaces (BCIs) in the elderly because of its compensatory role in neural circuits. However, it is difficult to decode FC by current machine learning techniques because of a lack of its physiological understanding. To investigate the suitability of FC in BCI for the elderly, we propose the decoding of lower- and higher-order FCs using a convolutional neural network (CNN) in six cognitive-motor tasks. The layer-wise relevance propagation (LRP) method describes how age-related changes in FCs impact BCI applications for the elderly compared to younger adults. Seventeen younger (24.5±2.7 years) and twelve older (72.5±3.2 years) adults were recruited to perform tasks related to hand-force control with or without mental calculation. CNN yielded a six-class classification accuracy of 75.3% in the elderly, exceeding the 70.7% accuracy for the younger adults. In the elderly, the proposed method increases the classification accuracy by 88.3% compared to the filter-bank common spatial pattern (FBCSP). LRP results revealed that both lower- and higher-order FCs were dominantly overactivated in the prefrontal lobe depending on task type. These findings suggest a promising application of multi-order FC with deep learning on BCI systems for the elderly.

As the world increasingly becomes more interconnected, the demand for safety and security is ever-increasing, particularly for industrial networks. This has prompted numerous researchers to investigate different methodologies and techniques suitable for intrusion detection systems (IDS) requirements. Over the years, many studies have proposed various solutions in this regard including signature-based and machine-learning (ML) based systems. More recently, researchers are considering deep learning (DL) based anomaly detection approaches. Most proposed works in this research field aimed to achieve either one or a combination of high accuracy, considerably low false alarm rates (FARs), high classification specificity and detection sensitivity, achieving lightweight DL models, or other ML and DL-related performance measurement metrics. In this study, we propose a novel method to convert a raw dataset to an image dataset to magnify patterns. Based on this we devise an anomaly detection for IDS using a lightweight convolutional neural network (CNN) that classifies denial of service and distributed denial of service. The proposed methods were evaluated using a modern dataset, CSE-CIC-IDS2018, and a legacy dataset, NSL-KDD. We have also applied a combined dataset to assess the generalization of the proposed model across various datasets. Our experimental results have demonstrated that the proposed methods achieved high accuracy and considerably low FARs with high specificity and sensitivity. The resulting loss and accuracy curves have also demonstrated the excellent generalization of the proposed lightweight CNN model, effectively avoiding overfitting. This holds for both the modern and legacy datasets, including their mixed version.

The most frequent, noticeable, and frequent natural calamity in the karakoram region is landslides. Extreme landslides have occurred frequently along Karakoram highway, particularly during the monsoon, causing a major loss of life and property. Therefore, it was necessary to look for a solution to increase growth and vigilance in order to lessen losses related to landslides caused by natural disasters. By utilizing contemporary technologies, an early warning system might be developed. Artificial neural networks (ANNs) are widely used nowadays across many industries. This paper's major goal is to provide new integrative models for assessing landslide susceptibility in a prone area of north of Pakistan. To do this, the training of an artificial neural network (ANN) is supervised using metaheuristic and Bayesian techniques: particle swarm optimization algorithm (PSO), Genetic algorithm (GA), Bayesian optimization Gaussian process (BO_GP), and Bayesian optimization Gaussian process (BO_TPE). 304 previous landslides and the eight most prevalent conditioning elements combine to form a geographical database. The models are hyper-parameter optimized, and the best ones are employed to generate the susceptibility maps. The area under the receiving operating characteristic curve (AUROC) accuracy index found demonstrated that the maps produced by both Bayesian and metaheuristic algorithms are highly accurate. The effectiveness and efficiency of applying artificial neural networks (ANNs) for landslide mapping, susceptibility analysis, and forecasting are studied in this research it’s observed from experimentation that the performance differences for GA, BO_GP, and PSO compared to BO_TPE are relatively small, ranging from 0.3166% to 1.8399%. This suggests that these techniques achieved comparable performance to BO_TPE in terms of AUC. However, it's important to note that the significance of these differences can vary depending on the specific context and requirements of the ML task. Additionally in this study, we explore eight feature selection algorithms to determine the geospatial variable importance for landslide susceptibility mapping along the KKH. The algorithms considered include Information Gain, Gain Ratio, OneR Classifier, Subset Evaluators, Principal Components, Relief Attribute Evaluator, Correlation, and Symmetrical Uncertainty. These algorithms enable us to evaluate the relevance and significance of different geospatial variables in predicting landslide susceptibility. By applying these feature selection algorithms, we aim to identify the most influential geospatial variables that contribute to landslide occurrences along the KKH. The algorithms encompass a diverse range of techniques, such as measuring entropy reduction, accounting for attribute bias, generating single rules, evaluating feature subsets, reducing dimensionality, and assessing correlation and information sharing. The findings of this study will provide valuable insights into the critical geospatial variables associated with landslide susceptibility along the KKH. These insights can aid in the development of effective landslide mitigation strategies, infrastructure planning, and targeted hazard management efforts. Additionally, the study contributes to the field of geospatial analysis by showcasing the applicability and effectiveness of various feature selection algorithms in the context of landslide susceptibility mapping.

Time series data having low signal-to-noise ratio, non-stationarity and non-linearity are commonly seen in high-frequency stock trading, where the objective is to increase the likelihood of profit by taking advantage of tiny discrepancies in prices and trading on them quickly and in huge quantities. For this purpose, it is essential to apply a trading method that is capable of fast and accurate prediction from such time series data. In this paper, we develop an online time series forecasting method for high-frequency trading (HFT) by integrating three neural network deep learning models, i.e., Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU) and Transformer; and we abbreviate the new method to online LGT or O-LGT. The key innovation underlying our method is its efficient storage management, which enables super-fast computing. Specifically, when computing the forecast for the immediate future time, we use only the output calculated from the previous trading data (rather than the previous trading data themselves) together with the current trading data. Thus, the computing involves updating only the current data into the process. We evaluate the performance of O-LGT by analyzing the high-frequency Limit Order Book (LOB) data from the China market. It shows that our model in most cases achieves similar speed with much higher accuracy than the conventional fast supervised learning models for HFT. However, with a slight sacrifice in accuracy, O-LGT is approximately 40 times faster than the existing high-accuracy neural network models for the LOB data in China market.

River bed materials serve multiple environmental functions as a habitat for aquatic invertebrates and fishes. At the same time, the particle size of the bed material reflects the tractive force of the flow regime in a flood and provides useful information for flood control. The traditional river bed particle size surveys, such as sieving, require time and labor to investigate river bed materials. The authors proposed a method to classify aerial images taken by unmanned aerial vehicle (UAV) using convolutional neural networks (CNN), our previous study showed that terrestrial riverbed material could be classified with high accuracy. In this study, we attempted to classify riverbed materials distributed in shallow waters where the bottom can be seen from UAVs. After training the CNN to classify the images with the same grain size as being in the same class even if the surface flow types taken overlapping the riverbed material were different, the total accuracy reached 90.3%. Moreover, the proposed method was applied to the wide-ranging area to determine the distribution of the particle size. In parallel, the microtopography was surveyed using Lidar-UAV, and the relationship between the microtopography and particle size distribution was discussed. In the steep section, coarse particles were distributed and formed a rapid. Fine particles were deposited on the upstream side of those rapids, where the slope had become gentler due to the damming. There was good agreement between the microtopographical trends and the grain size distribution.

Early prediction of thermal loads plays an essential role in analyzing energy-efficient buildings' energy performance. On the other hand, stochastic algorithms have recently shown high proficiency in dealing with this issue. These are the reasons that this work is dedicated to evaluating an innovative hybrid method for predicting the cooling load (CL) in buildings with residential usage. The proposed model is a combination of artificial neural networks and stochastic fractal search (SFS-ANN). Two benchmark algorithms, namely the grasshopper optimization algorithm (GOA) and firefly algorithm (FA), are also considered to be compared with the SFS. The non-linear effect of eight independent factors on the CL is analyzed using each model's optimal structure. Evaluation of the results outlined that all three metaheuristic algorithms (with more than 90 % correlation) can adequately optimize the ANN. In this regard, this tool's prediction error declined by nearly 23, 18, and 36 % by applying the GOA, FA, and SFS techniques. Also, all used accuracy criteria indicated the superiority of the SFS over the benchmark schemes. Therefore, it is inferred that utilizing the SFS along with ANN provides a reliable hybrid model for the early prediction of CL.

This paper presents a novel architecture for detecting mathematical formulas in document images, which is an important step for reliable information extraction in several domains. Recently, Cascade Mask R-CNN networks have been introduced to solve object detection in computer vision. In this paper, we suggest a couple of modifications to the existing Cascade Mask R-CNN architecture: First, the proposed network uses deformable convolutions instead of conventional convolutions in the backbone network to spot areas of interest better. Second, it uses a dual backbone of ResNeXt-101, having composite connections at the parallel stages. Finally, our proposed network is end-to-end trainable. We evaluate the proposed approach on the ICDAR-2017 POD and Marmot datasets. The proposed approach demonstrates state-of-the-art performance on ICDAR-2017 POD at a higher IoU threshold with an f1-score of 0.917, reducing the relative error by 7.8%. Moreover, we accomplished correct detection accuracy of 81.3% on embedded formulas on the Marmot dataset, which results in a relative error reduction of 30%.

One of the critical tools for early detection and subsequent evaluation of the incidence of lung diseases is chest radiography. At a time when the speed and reliability of results, especially for COVID-19 positive patients, is important, the development of applications that would facilitate the work of untrained staff involved in the evaluation is also crucial. Our model takes the form of a simple and intuitive application, into which you only need to upload X-rays: tens or hundreds at once. In just a few seconds, the physician will determine the patient's diagnosis, including the percentage accuracy of the estimate. While the original idea was a mere binary classifier that could tell if a patient was suffering from pneumonia or not, in this paper we present a model that distinguishes between a bacterial disease, a viral infection, or a finding caused by COVID-19. The aim of this research is to demonstrate whether pneumonia can be detected or even spatially localized using a uniform, supervised classification.

Making sustainable business models, especially in manufacturing, is a key approach for organizations to adopt in order to create more sustainable processes. New technologies are needed to implement new policies that can minimize the economic, environmental, and social consequences of industrial activity. Sustainable maintenance is decisive to guarantee the availability, reliability, and safety of assets, surpassing traditional maintenance policies and embracing the principles of the circular economy. The main objective of Maintenance 4.0 is to increase the lifespan and availability of the equipment by preventing unplanned downtime, reducing planned shutdowns, and improving safety. This research focuses on the integration of advanced Industry 4.0 tools with conventional monitoring and control techniques in process industries. The article suggests that while complex neural networks might provide good predictions, simpler solutions can generalize well and have lower prediction errors because they can capture the underlying relationships between variables in a specific asset better. The combination of maintenance function, Industry 4.0 tools, and advanced statistical solutions to obtain sustainable processes can facilitate organizational change and aid decision-making in equipment management. These solutions seek to reduce interruptions and expenses through the unnecessary utilization of labor and spare parts. All these considerations improve communication and data processing, resulting in improved maintenance of a hydrogen compressor in the petrochemical industry.

Memristors are state-of-the-art, nano-sized, two-terminal, passive electronic elements with very good switching and memory characteristics. Owing to their very low power usage and a good compatibility to the existing CMOS ultra-high-density integrated circuits and chips, they are po-tentially applicable in artificial and spiking neural networks, memory arrays, and many other de-vices and circuits for artificial intelligence. In this paper, a complete electronic realization of ana-log circuit model of modified neural net with memristor-based synapses and transfer function with memristors and MOS transistors in LTSPICE is offered. Each synaptic weight is realized by only one memristor, providing enormously reduced circuit complexity. The summing and scaling implementation is founded on op-amps and memristors. The logarithmic-sigmoidal activation function is based on a simple scheme with MOS transistors and memristors. The functioning of the suggested memristor-based neural network for pulse input signals is evaluated both analytically in MATLAB-SIMULINK and in LTSPICE environment. The obtained results are compared one to another and are successfully verified. The realized memristor-based neural network is an im-portant step towards the forthcoming design of complex memristor-based neural networks for ar-tificial intelligence, for implementation in very high-density integrated circuits and chips.

This paper introduces a novel ensemble adjustment Kalman filter (EAKF) that integrates a machine-learning approach. The approach employs nonlinear variable relationships established by a deep neural network (DNN) during the analysis stage of the EAKF. This process nonlinearly projects observation increments into the state variable space. The newly developed DNN-EAKF algorithm can be applied to coupled data assimilation using coupled ocean-atmosphere models. It enhances cross-component updates in strongly coupled data assimilation (SCDA) by diminishing errors in estimating cross-component error covariance arising from insufficient ensemble members, thereby improving the SCDA analysis. This paper employs a conceptual model to conduct twin experiments, validating the DNN-EAKF’s capability to utilize cross-component observation information in SCDA effectively. The approach is anticipated to offer insights for future methodological integrations of machine learning and data assimilation and provide methods for SCDA applications in coupled general circulation models.

Lithium-ion batteries serve as the primary sources of power for electric vehicles (EVs) and hybrid electric vehicles (HEVs). For vehicle applications, battery management systems (BMSs) are nec-essary to protect lithium-ion batteries from overheating and to ensure optimum vehicle perfor-mance. Our approach to developing a BMS was based on recent advances in the application of phase field models for lithium-ion batteries. In particular, our reduced-order model (ROM) uti-lized a dataset generated from the COMSOL® Multiphysics simulation of the Cahn–Hilliard equation for a single particle of a lithium iron phosphate (LiFePO4) cathode: an example of using a reduced-order model (ROM) based on a single-particle model (SPM). The main innovation of our ROM is that the SPM is fully coupled to a heat transfer model at the battery cell level. We utilized principal component analysis to identify a lower-order model that could reproduce the battery’s voltage and temperature response for ambient temperatures ranging from 253 to 298 K and for discharge rates ranging from 1 C to 20.5 C. The reduced-order dataset was then fitted to the ex-perimental data for an A123 Systems 26650 2.3 Ah cylindrical battery using deep neural network (DNN) regression.

Accurate evaluation of health status of lithium-ion batteries must be deemed as of great significance, insofar as utility and safety of batteries are of concern. Lithium plating, in particular, is notoriously known to be a chemical reaction that can cause deterioration in, or even fatal hazards to, the health of lithium-ion batteries. Electrochemical impedance spectroscopy (EIS), which has distinct advantages such as fast or non-destructive over its competitors, suffices in detecting lithium plating and thus has been attracting increasing attention in the field of battery management, but its ability of assessing quantitatively the degree of lithium plating remains largely unexplored hitherto. On this point, this work seeks to narrow that gap, by proposing an EIS-based method that can quantify the degree of lithium plating. The core conception is to eventually circumvent the reliance on state-of-health measurement, and use instead the impedance spectrum, to acquire an estimate on battery capacity loss. To do so, the effects of solid electrolyte interphase formation and lithium plating on battery capacity must be first decoupled, so that the mass of lithium plating can be quantified. Then, based on an impedance spectrum measurement, the parameters of the fractional equivalent circuit model (ECM) of the battery can be identified. These fractional ECM parameters are received as inputs by an artificial neural network, which is tasked to establish a correspondence between the model parameters and the mass of lithium plating. The empirical part of the work revolves around the data collected from an aging experiment, and the validity of the proposed method is truthfully attested by dismantling the batteries, which is otherwise not needed during the actual uptake of the method.

Agriculture is one of the indispensable fields for the survival of mankind. Potatoes also have a significant role in the field of agriculture. The quality and quantity of potatoes are significantly impacted by several diseases, such as early blight and late blight, and manual interpretation of these leaf diseases is time-consuming and inconvenient. Fortunately, leaf appearance is used to detect diseases in potato plants. Productivity will considerably rise if the infections are detected early. Different types of image processing methods and machine learning methods are used for early recognition of those diseases from leaf images so that the product losses are decreased significantly. For the incredible and marvelous performance of CNN, it is the most popular deep learning method that is used immensely for the recognition of leaf diseases from images. Several pretrained deep learning models, such as VGG16, ResNet50, InceptionV3, MobileNetV2, Xception, and a deep learning model developed using CNN, are employed for potato leaf disease classification and recognition on the same dataset. The transfer learning technique is applied to the pretrained model and the data augmentation technique is applied to the Proposed CNN model for potato disease classification from leaf samples. Compared to pretrained models, the proposed CNN model offers the lowest loss and highest accuracy for potato leaf disease detection while using fewer parameters and layers. It achieves the best performance with a test accuracy of 99.33% compared to other Pretrained models used in the diagnosis of potato leaf disease.

This research project aimed to address the growing concern about methane emissions from seaweed by developing a Convolutional Neural Network (CNN) model capable of accurately predicting these emissions. The study used PANDAS to read and analyze the dataset, incorporating statistical measures like mean, median, and standard deviation to understand the dataset. The CNN model was trained using the ReLU activation function and mean absolute error as the loss function. Model performance was evaluated through MAPE graphs, comparing the mean absolute percentage error (MAPE) between training and validation sets and between true and predicted emissions, and analyzing trends in yearly greenhouse gas emissions. The results demonstrated that the CNN model achieved a high level of accuracy in predicting methane emissions, with a low MAPE between the expected and actual values. This approach should enhance our understanding of methane emissions from Sargassum, contributing to more accurate environmental impact assessments and effective mitigation strategies.

Spherical harmonic (SH) interpolation is a commonly used method to spatially up-sample sparse Head Related Transfer Function (HRTF) datasets to denser HRTF datasets. However, depending on the number of sparse HRTF measurements and SH order, this process can introduce distortions in high frequency representation of the HRTFs. This paper investigates whether it is possible to restore some of the distorted high frequency HRTF components using machine learning algorithms. A combination of Convolutional Auto-Encoder (CAE) and Denoising Auto-Encoder (DAE) models is proposed to restore the high frequency distortion in SH interpolated HRTFs. Results are evaluated using both Perceptual Spectral Difference (PSD) and localisation prediction models, both of which demonstrate significant improvement after the restoration process.

Emergency control load-shedding is a key solution to prevent blackouts in the power system. This paper proposed a new model of emergency controls load shedding based on the fast identification of the unstable state of the power system. K-means clustering algorithm divided the instability mode into the clusters. The results of analysis of this cluster were used as the basis for classification control. Building load shedding strategies is consisted of the pre-designed rules based on AHP algorithm. When the recognition of the power system “instability” is detected, the signal of load shedding control is triggered immediately, therefore the decision time is greatly shortened comparing to the traditional methods. The effectiveness of the proposed method was tested on the IEEE 39-bus to overcome the limitations of the last traditional methods.

Photoelectric smoke detectors are the most cost-effective devices for very early fire alarms. However, due to different light intensity response values for different fire smoke and interference from interferential aerosols, they have a high false alarm rate, which limits their popularity in Chinese homes. To address these issues, an embedded spatial-temporal convolutional neural network (EST-CNN) model is proposed for real fire smokes identification and aerosols (fire smokes and interferential aerosol) classification. EST-CNN consists of three modules including information fusion, scattering feature extraction, and aerosol classification. Moreover, a two dimensional spatial-temporal scattering (2D-TS) matrix is designed to fuse the scattered light intensities in different channels and adjacent time slices, which is the output of the information fusion module and the input of the scattering feature extraction module. EST-CNN is trained and tested with experimental data measured on the established fire test platform using the developed dual-wavelength dual-angle photoelectric smoke detector. The optimal network parameters are selected through extensive experiments resulting in an average classification accuracy of 95.6% for different aerosols with only 66 kB network parameters. The experimental results demonstrate the feasibility of the designed EST-CNN model to be directly installed in existing commercial photoelectric smoke detectors to realize aerosol classification.

A block-diagonal fuzzy neural network for short-term load forecasting is proposed. DBD-FELF consists of fuzzy rules with consequent parts that are neural networks with internal recurrence. These networks have a hidden layer which consists of pairs of neurons with feedback connections between them. The overall fuzzy model partitions the input space in partially overlapping fuzzy regions, where the recurrent neural networks of the respective rules operate. The partition of the input space and determination of the fuzzy rule base is performed by use of Fuzzy C-Means clustering algorithm and the RENNCOM constrained optimization method is applied for consequent parameter tuning. The electric load time-series of the Greek power system is examined, and hourly-based forecasting for the whole year is performed. The performance of DBD-FELF is tested via extensive experimental analysis and the results are promising, since an average percentage error of 1.18% is attained, along with an average yearly absolute error of 76.2 MW. Moreover, DBD-FELF is compared with Deep Learning, fuzzy and neurofuzzy rivals, such that its particular attributes are highlighted.

Accurate forecasts of ocean waves energy can not only reduce costs for investment but it is also essential for management and operation of electrical power. This paper presents an innovative approach based on the Long Short Term Memory (LSTM) to predict the power generation of an economical wave energy converter named “Searaser”. The data for analyzing is provided by collecting the experimental data from another study and the exerted data from numerical simulation of searaser. The simulation is done with Flow-3D software which has high capability in analyzing the fluid solid interactions. The lack of relation between wind speed and output power in previous studies needs to be investigated in this field. Therefore, in this study the wind speed and output power are related with a LSTM method. Moreover, it can be inferred that the LSTM Network is able to predict power in terms of height more accurately and faster than the numerical solution in a field of predicting. The network output figures show a great agreement and the root mean square is 0.49 in the mean value related to the accuracy of LSTM method. Furthermore, the mathematical relation between the generated power and wave height was introduced by curve fitting of the power function to the result of LSTM method.

Functional Connectivity analysis using Electroencephalography signals is common. The EEG signals are converted to networks by transforming the signals into a correlation matrix and analyzing the resulting networks. Here, four learning models, namely, Logistic Regression, Random Forest, Support Vector Machine, and Recurrent Neural Networks, are implemented on the correlation matrix data to classify them either on their psychometric assessment or the effect of therapy; The EEG data is trail-based/event-related. The classifications based on RNN provided higher accuracy( 74-88%) than the other three models( 50-78%). Instead of using individual graph features, a correlation matrix provides an initial test of the data. When compared with the time-resolved correlation matrix, it offered a 4-5% higher accuracy. The time-resolved correlation matrix is better suited for dynamic studies here; it provides lower accuracy when compared to the correlation matrix, a static feature.

Deep learning has been applied in many areas that have had a significant impact on applications that improves real-life challenges. The success of deep learning in a wide range of areas is due in part to the use of activation functions, which are particularly effective at solving non-linear problems. Activation functions are a key focus for researchers in artificial intelligence who aim to improve the performance of neural networks. This article provides a comprehensive explanation and comparison of different activation functions, with a focus on the arc tangent and its variations specifically. The paper presents experimental results that show that variations of the arc tangent using irrational numbers such as pi, the golden ratio, and Euler’s number, as well as a self-arctan function, produce promising results. Since we experimented with promising activation functions on two different problems, and datasets, we reached a result that different irrationals work well for different problems. In other words, arctan ϕ gives the best results mostly for multiclass classification and arctan e gives the best results for time series prediction problems. The paper focuses on a multiclass classification problem applied to the Reuters Newswire dataset and a time-series prediction problem on Türkiye energy trade value to show the impacts of activation functions.

Stock price prediction has long been a critical area of research in financial modeling. The inherent complexity of financial markets, characterized by both short-term fluctuations and long-term trends, poses significant challenges in accurately capturing underlying patterns. While Long Short-Term Memory (LSTM) networks have shown strong performance in short-term stock price prediction, they struggle with effectively modeling long-term dependencies. In this paper, we propose an advanced stock price prediction model based on the Extended Long Short-Term Memory (xLSTM) algorithm, designed to enhance predictive accuracy over both short and long-term periods. We conduct extensive experiments by building and evaluating models based on xLSTM and LSTM architectures for multiple stocks. Our results demonstrate that the xLSTM model consistently outperforms the LSTM model across all stocks and time horizons, with the performance gap widening as the prediction period extends. The observations underscore the superior capability of the xLSTM-based model to capture long-term patterns in financial data, offering a promising approach for more accurate and reliable stock price predictions.

A reinforcement learning algorithm is proposed to improve the accuracy of short-term load forecasting (STLF) in this article. The proposed model integrates radial basis function neural network (RBFNN), support vector regression (SVR), and adaptive annealing learning algorithm (AALA). In the proposed methodology, firstly, the initial structure of RBFNN is determined by using SVR. Then, an AALA with time-varying learning rates is used to optimize the initial parameters of SVR-RBFNN (AALA-SVR-RBFNN). In order to overcome the stagnation for searching optimal RBFNN, a particle swarm optimization (PSO) is applied to simultaneously find promising learning rates in AALA. Finally, the short-term load demands are predicted by using the optimal RBFNN. The performance of the proposed methodology is verified on the actual load dataset from Taiwan Power Company (TPC). Simulation results reveal that the proposed AALA-SVR-RBFNN can achieve a better load forecasting precision as compared to various RBFNNs.

During planetary exploration mission operations, one of the key responsibilities of the instrument teams to determine data viability for subsequent analysis. During the 2019 CanMoon Lunar Sample Return Analogue Mission, the Lead Raman Specialist manually examined each spectra to provide quality assurance/validation. This non-trivial process requires years of experience to complete accurately. With the proven efficacy of Convolutional Neural Networks (CNNs) in classification tasks, and the increased use of automation and control loops on planetary space platforms for navigation and science targeting, an opportunity presents itself to approach this validation problem utilising CNNs. We present the Generalised Raman Validation Network (GRaVN), an neural network focused specifically on extracting the generalised structure of Raman spectra for quality assurance/validation. This work demonstrates the viability of utilising a CNN network in validation activities for Raman spectroscopy. Utilising only two hidden layers, a configuration was developed that provided good levels of accuracy on a manually curated dataset. This indicates that such a system could be useful as part of an autonomous control loop during planetary exploration activities.

Convolutional Neural Networks (CNNs) are often criticized for their lack of transparency, acting as 'black boxes' in decision-making, a challenge compounded by class imbalance in ECG datasets, which limits their clinical application. This study introduces a CNN-based ECG signal classification model that enhances interpretability and addresses class imbalance through the Synthetic Minority Over-sampling Technique (SMOTE). The model also integrates Uniform Manifold Approximation and Projection (UMAP) for dimensionality reduction and SHAP value analysis, facilitating the visualization of decision boundaries and the assessment of feature contributions. Our evaluation using the MIT-BIH Arrhythmia Database highlights the model's high performance, with accuracy and precision nearing 1.00 for Normal ECG (NOR), Left Bundle Branch Block (LBBB), Right Bundle Branch Block (RBBB), and Ventricular Premature Beat (PVC) in the six-class task. In the ten-class task, the model demonstrated robustness, particularly with an accuracy of 0.9846, precision of 0.9783, recall of 0.9736, and F1 score of 0.9760 in Pacemaker Fusion Beat (PFHB), supported by an AUC of 0.9999 and AP of 0.9885. These results underscore the model's efficacy in cardiac rhythm recognition and resilience to class imbalance. Future research will explore sophisticated model architectures and feature extraction methods to enhance the model's generalization and clinical applicability for early heart disease diagnosis and personalized treatment.

The integration of deep learning methods, especially Convolutional Neural Networks (CNN), and Synthetic Aperture Radar Automatic Target Recognition (SAR ATR) has been widely deployed in the field of radar signal processing. Nevertheless, these methods are frequently regarded as black-box models due to the limited visual interpretation of their internal feature representation and parameter organization. In this paper, we propose an innovative approach named C-RISE, which builds upon the RISE algorithm to provide a post-hoc interpretation technique for black-box models used in SAR Images Target Recognition. C-RISE generates saliency maps that effectively visualize the significance of each pixel. Our algorithm outperforms RISE by clustering masks that capture similar fusion features into distinct groups, enabling more appropriate weight distribution and increased focus on the target area. Furthermore, we employ Gaussian blur to process the masked area, preserving the original image structure with optimal consistency and integrity. C-RISE has been extensively evaluated through experiments, and the results demonstrate superior performance over other interpretation methods based on perturbation when applied to neural networks for SAR image target recognition. Furthermore, our approach is highly robust and transferable compared to other interpretable algorithms, including white-box methods.

Pedestrian object detection plays a significant role in intelligent systems such as intelligent traffic and monitoring. Traditional machine learning methods on pedestrian detection have shown various drawbacks, e.g., low accuracy, slow speed, etc. The Convolutional Neural Network (CNN) based object detection algorithms have demonstrated remarkable advantages in the field of pedestrian detection. However, the mainstream CNNs still face the problems of slow speed and low detection accuracy, especially on small and occluded targets from aerial perspective. In this paper, we propose Multi-Scale Attention YOLO (MSA-YOLO) detection algorithm to address the above issues. MSA-YOLO includes a Squeeze, Excitation and Cross Stage Partial (SECSP) channel attention module for CNNs to extract richer pedestrian features with a small number of extra parameters. It also contains a multi-scale prediction module to capture the information among different pedestrian scales, which can recognize the small objects with higher accuracy and significantly reduce the missed detection. To sufficiently evaluate our proposed model, we manually collect and annotate a new benchmark dataset, Aerial Pedestrian Dataset, which has much more sample annotations, features, scenes and image view angles than the existing benchmark datasets. In addition, the images in our dataset have higher resolution than most of benchmark pedestrian detection datasets, which can provide more detailed features of pedestrians and thus improve the model performance. When tested on Aerial Pedestrian Dataset, our proposed MSA-YOLO algorithm significantly outperforms the most used baseline models with almost the same model size. This shows the efficiency of our proposed model. (The code and new dataset will be released to the public later.)

— In a smart healthcare system," Human Activity Recognition (HAR)" is considered as an efficient approach in pervasive computing from activity sensor readings. The "Ambient Assisted Living (AAL)" in the home or community helps the people to provide independent care and enhanced living quality. However, many AAL models are restricted to multiple factors that include both the computational cost and system complexity. Moreover, the HAR concept has more relevance because of its applications, such as content-based video search, sports play analysis, crowd behavior prediction systems, patient monitoring systems, and surveillance systems. This paper attempts to implement the HAR system using a popular deep learning algorithm, namely "Recurrent Neural Network (RNN)" with the activity data collected from smart activity sensors over time, and it is publicly available in the "UC Irvine Machine Learning Repository (UCI)". The proposed model involves three processes: (1) data collection, (b) optimal feature learning, and (c) activity recognition. The data gathered from the benchmark repository was initially subjected to optimal feature selection that helped to select the most significant features. The proposed optimal feature selection method is based on a new meta-heuristic algorithm called "Colliding Bodies Optimization (CBO)". An objective function derived from the recognition accuracy has been used for accomplishing the optimal feature selection. The proposed model on the concerned benchmark dataset outperformed the conventional models with enhanced performance.

Skin cancer is one of the widespread diseases that typically develop on the skin due to continuous exposure to sunlight. Although cancer can appear on any part of the human body, skin cancer reports account for over half of all cancer occurrences worldwide. There are substantial obstacles to the precise diagnosis and classification of skin lesions because of the morphological variety and indistinguishable characteristics across skin malignancies. Recently, Deep Learning models have been used in the field of image-based lesion diagnosis, and it has demonstrated diagnostic efficiency on par with that of dermatologists. To increase classification efficiency and accuracy for skin lesions, a cutting-edge multi-layer deep Convolutional Neural Network (CNN) termed SkinLesNet has been built in this study. The ResNetV50 and VGG16 models have been carefully compared to review the performance of the proposed model. The dataset used in this study, PAD-UFES-20, contains 1314 samples in total and includes three common forms of skin lesions. The proposed approach, SkinLesNet, significantly outperforms the well-known compared models in the given conditions

An approach to the implementation of neural network for real-time cryptographic data protection with symmetric keys oriented on embedded systems is presented. This approach is valuable especially for onboard communication systems in unmanned aerial vehicles (UAV) because of suitability to hardware implementation. In this study we evaluate the possibility to build such a system in hardware implementation at FPGA. The proposed integrated neural network approach for real-time cryptographic data protection was based on theoretical foundations of neural network cryptographic data protection, new algorithms and structures of neural network data encryption and decryption, modern hardware components with programmable structure. The development and implementation of the on-board neural network system for cryptographic data protection in real-time are based on the following principles: variable composition of equipment; modularity; conveyorization and spatial parallelism; hardware and software modification and making them suitable for data encryption and decryption. The tabular-algorithmic method for calculation of the scalar product has been improved that makes it possible to perform fast calculation of the scalar product of fixed-point and floating-point input data by bringing them to the largest common order of weights and forming tables of macro-partial products. Components of neural network cryptographic data encryption and decryption have been developed on the processor core supplemented by specialized scalar product calculation modules. The specialized hardware for neural network cryptographic data encryption was developed using VHDL for equipment programming in the Quartus II development environment ver. 13.1 and the appropriate libraries, and implemented on the basis of the FPGA EP3C16F484C6 Cyclone III family.

Deep learning has been applied in many areas that have had a significant impact on applications that improves real-life challenges. The success of deep learning in a wide range of areas is due in part to the use of activation functions, which are particularly effective at solving non-linear problems. Activation functions are a key focus for researchers in artificial intelligence who aim to improve the performance of neural networks. This article provides a comprehensive explanation and comparison of different activation functions, with a focus on the arc tangent and its variations specifically. The paper presents experimental results that show that variations of the arc tangent using irrational numbers such as pi, the golden ratio, and Euler’s number, as well as a self-arctan function, produce promising results. Since we experimented with promising activation functions on two different problems, and datasets, we reached a result that different irrationals work well for different problems. In other words, arctan ϕ gives the best results mostly for multiclass classification and arctan e gives the best results for time series prediction problems. The paper focuses on a multi-class classification problem applied to the Reuters Newswire dataset and a time-series prediction problem on Türkiye energy trade value to show the impacts of activation functions.

Total cloud cover (TCC) retrieval from ground-based optical imagery is a problem being tackled by a few generations of researchers. The number of human-designed algorithms for the estimation of TCC grows every year. However, there is not very much progress in terms of quality, mostly due to the lack of systematic approach to the design of the algorithms, to the assessment of their generalization ability, and to the assessment of the TCC retrieval quality. In this study, we discuss the optimization nature of data-driven schemes for TCC retrieval. In order to compare the algorithms, we propose the framework for the assessment of the algorithms characteristics. We present several new algorithms that are based on deep learning techniques: a model for outliers filtering, and a few models for TCC retrieval from all-sky imagery. For training and assessment of data-driven algorithms of this study, we present the Dataset of All-Sky Imagery over the Ocean (DASIO) containing over one million of all-sky optical images of visible sky dome taken in various regions of the World Ocean. The research campaigns contributed to DASIO collection took place in the Atlantic ocean, the Indian ocean the Red and Mediterranean seas, and also in the Arctic ocean. Optical imagery collected during these missions are accompanied by standard meteorological observations of cloudiness characteristics made by experienced observers. We assess the generalization ability of the presented models in several scenarios that differ in terms of the regions selected for the train and validation subsets. As a result, we demonstrate that our models based on convolutional neural networks deliver superior quality compared to all previously published schemes. As a key result, we demonstrate considerable drop of the ability to generalize the training data in case of strong covariate shift between training and validation subsets of imagery which may take place in case of region-aware subsampling.

In this paper, we demonstrate that fixed-weight layers, generated from random distribution or logistic functions, can effectively extract significant features from input data, resulting in high accuracy on a variety of tasks, including Fisher’s iris, Wisconsin Breast Cancer, and MNIST datasets. We have observed that logistic functions yield high accuracy with less dispersion in results. We have also assessed the precision of our approach under conditions of minimizing spikes number generated in the network, it is a practically useful for reducing energy consumption in spiking neural networks. Our findings reveal that the proposed method demonstrates a highest accuracy on Fisher’s iris and MNIST datasets with logistic regression. Furthermore, they surpass the accuracy of the conventional (non-spiking) approach using logistic regression in the case of Wisconsin Breast Cancer. We have also investigated the impact of non-stochastic spike generation on accuracy.

This study aims to evaluate a new approach in modeling gully erosion susceptibility based on deep learning neural network (DLNN) model, ensemble Particle swarm optimization (PSO) algorithm with DLNN (PSO-DLNN) and comparing these approaches with common artificial neural network (ANN) and support vector machine (SVM) models in Shiran watershed, Iran. For this purpose, 13 independent variables affecting gully erosion susceptibility in the study area, including altitude, slope, aspect, plan curvature, profile curvature, drainage density, distance from river, land use, soil, lithology, rainfall, , stream power index (SPI), topographic wetness index (TWI), were prepared. Also, 132 gully erosion locations were identified during field visits. Data for modeling were divided into two categories of training (70%) and testing (30%). Receiver operating characteristic (ROC) parameters including sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) and area under curve (AUC) were used to evaluate the performance of the models. The results showed that, the AUC values from ROC with considering testing datasets of PSO-DLNN is 0.89 and which is associated with superb accuracy. Rest of the models also associated with optimal accuracy and near about PSO-DLNN model; the AUC values from ROC of DLNN, SVM and ANN for testing datasets are 0.87, 0.85 and 0.84 respectively. The PSO algorithm has updated and optimized the weights of DLNN model, and as a result, the efficiency of this model in predicting gully erosion susceptibility has increased. Therefore, it can be concluded that the use of DLNN model and its ensemble with PSO algorithm can be used as a novel and practical method in predicting the susceptibility of gully erosion that helps planners and managers in managing and reducing the risk of this phenomenon.

The problem of optimal scheduling in a system with parallel queues and a single server has been extensively studied in the queueing theory. However, such systems have mostly been studied by assuming homogeneous attributes of arrival and service processes, or in heterogeneous case a Markov random process was usually assumed. In this paper, we explore the possibility of combining simulation and neural network paradigms for optimal scheduling in a heterogeneous system with an arbitrary inter-arrival and service time distributions. The service of any queue by the server is exhaustive. Further routing of the server after the queue has been emptied is as instructed by the trained neural network. Any change in the queue that currently has a server is accompanied by a switching cost. The simulated annealing optimization algorithm which is used to optimize the parameters of the neural network was adapted for numerically evaluated average cost function. A Markov decision problem was formulated for the corresponding markovian queueing system to verify the results of solving the optimization problem. In addition to comparing the actual values of the average cost, the parameter values of the neural network trained on the optimal control policy of the Markov model are also used to determine the domains of parameters required for the simulated annealing defined on a finite discrete domain. Numerical analysis shows the effectiveness of this approach. Moreover, a comparison of the results for different distributions illustrates the insensitivity of the optimal control policy to the shape of the distributions for the same two first moments. Thus, the optimal control policy for an exponential model can be used as a suboptimal one for the general system.

With the automotive industry moving towards automated driving, sensing is increasingly important in enabling technology. The virtual sensors allow data fusion from various vehicle sensors and provide a prediction for measurement that is hard or too expensive to measure in another way or in the case of demand on continuous detection. In this paper, virtual sensing is discussed for the case of vehicle suspension control, where information about the relative velocity of the unsprung mass for each vehicle corner is required. The corresponding goal can be identified as a regression task with multi-input sequence input. The hypothesis is that the state-of-art method of Bidirectional Long-Short Term Memory (BiLSMT) can solve it. In this paper, a virtual sensor has been proposed and developed by training a neural network model. The simulations have been performed using an experimentally validated full vehicle model in IPG Carmaker. Simulations provided the reference data which was used for Neural Network (NN) training. The extensive dataset covering 26 scenarios has been used to obtain training, validation and testing data. The Bayesian Search was used to select the best neural network structure using root mean square error as a metric. The best network is made of 167 BiLSTM, 256 fully connected hidden units and 4 output units. Error histograms and spectral analysis of the predicted signal compared to the reference signal are presented. The results demonstrate the good applicability of neural network-based virtual sensors to estimate vehicle unsprung mass relative velocity.

The importance of this study may be defined by using the smart techniques to earned value indicators of residential buildings projects in Republic of Iraq, only one development intelligent forecasting model was presented to predict Schedule Performance Index (SPI), Cost Performance Index (CPI), and To Complete Cost Performance Indicator (TCPI) are defined as the dependent. The approach is principally influenced by the determining numerous factors which effect on the earned value management, that involves Iraqi historical data. In addition, six independent variables (F1: BAC, Budget at Completion., F2: AC, Actual Cost., F3, A%, Actual Percentage., F4: EV, Earned Value. F5: P%, Planning Percentage., and F6: PV, Planning Value) were arbitrarily designated and satisfactorily described for per construction project. It was found that ANN has the capability to envisage the dust storm with a great accuracy. The correlation coefficient (R) has been 90.00%, and typical accuracy percentage has been 89.00%.

In this study, deep neural network (DNN) and transfer learning (TL) techniques were employed to predict the viscous resistance and wake distribution based on the positions of flow control fins (FCFs) applied to containerships of various sizes. Both methods utilized data collected through Computational Fluid Dynamics (CFD) analysis. The position of the flow control fin (FCF) and hull-form information were utilized as input data, and the output data included viscous resistance coefficients and components of propeller axial velocity. The base DNN model was trained and validated using a source dataset from a 1000 TEU containership. Grid search cross-validation technique was employed to optimize the hyperparameters of the base DNN model. Then, transfer learning was applied to predict the viscous resistance and wake distribution for containerships of varying sizes. To enhance the accuracy of feature prediction with a limited amount of dataset, learning rate optimization was conducted. Transfer learning involves retraining and reconfiguring the base DNN model, and the accuracy was verified based on the fine-tuning method of the learning model. The results of this study can provide hull designers for containerships with performance evaluation information by predicting wake distribution, without relying on CFD analysis.

Driver drowsiness is one of the leading causes of traffic accidents. This paper proposes a new method for classifying driver drowsiness using deep convolution neural networks trained by wavelet scalogram images of electrocardiogram (ECG) signals. Three different classes were de-fined for drowsiness based on video observation of driving tests performed in a simulator for manual and automated modes. The Bayesian optimization method is employed to optimize the hyperparameters of the designed neural networks, such as the learning rate and the number of neurons in every layer. To assess the results of the deep network method, Heart Rate Variability (HRV) data is derived from the ECG signals, some features are extracted from this data, and finally, random forest and k-nearest neighbors (KNN) classifiers are used as two traditional methods to classify the drowsiness levels. Results show that the trained deep network achieves balanced accuracies of about 77% and 79% in the manual and automated modes, respectively. However, the best obtained balanced accuracies using traditional methods are about 62% and 64%. We conclude that designed deep networks working with wavelet scalogram images of ECG signals significantly outperform KNN and random forest classifiers which are trained on HRV-based features.

Substantial evidence indicates that bisphenol A (BPA), a ubiquitous environmental chemical used in the synthesis of polycarbonate plastics and epoxy resins, can impair brain development. Clinical and epidemiological studies exploring potential connections between BPA and neurodevelopmental disorders in humans have repeatedly identified correlations between early BPA exposure and developmental disorders, like attention deficit/hyperactivity disorder and autism spectrum disorder. Investigations using invertebrate and vertebrate animal models have revealed that developmental exposure to BPA can impair multiple aspects of neuronal development, including neural stem cell proliferation and differentiation, synapse formation, and synaptic plasticity—neuronal phenotypes that are thought to underpin the fundamental changes in behavior associated neurodevelopmental disorders. Consistent with BPA-associated neuronal phenotypes, behavioral analyses of BPA-treated animals have shown significant impacts on behavioral endophenotypes related to neurodevelopmental disorders, including altered locomotor activity, learning and memory deficits, and anxiety-like behavior. To contextualize the correlations between BPA and neurodevelopmental disorders in humans, this review summarizes current literature reporting on the developmental neurotoxicity of BPA in laboratory animals, with an emphasis on neuronal phenotypes, molecular mechanisms, and behavioral outcomes. The collective works described here predominantly support the notion that gestational exposure to BPA should be regarded as a risk factor for neurodevelopmental disorders.

Indian cities are increasingly becoming susceptible to PM10 induced health effects which have become a matter of concern for the policymakers of the country. Air pollution is engulfing the comparatively smaller cities as the rapid pace of urbanization, and economic development seems not to lose steam ever. A review of air pollution of 28 cities of India, which includes tier-I, II, and III cities of India, found to have grossly violated both WHO and NAAQS standards in respect of acceptable daily average PM10 concentrations by a wide margin. Predicting the city level PM10 concentrations in advance and accordingly initiate prior actions is an acceptable solution to save the city dwellers from PM10 induced health hazards. Predictive ability of three models, linear MLR, nonlinear MLP (ANN), and nonlinear CART, for one day ahead PM10 concentration forecasting of tier-II Guwahati city, were tested with 2016-2018 daily average observed climate data, PM10, and gaseous pollutants. The results show that the non-linear algorithm MLP with feedforward backpropagation network topologies of ANN class, giving the best prediction value when compared with linear MLR and nonlinear CART model. ANN (MLP) approach, therefore, may be useful to effectively derive a predictive understanding of one day ahead PM10 concentration level and thus provide a tool to the policymakers for improving decision-making associated with air pollution and public health.

This paper proposes two max-pooling engines, named the RTB-MAXP engine and the CMB-MAXP engine, with a scalable window size parameter for FPGA-based convolutional neural network (CNN) implementation. The max-pooling operation for the CNN can be decomposed into two stages, i.e., a horizontal axis max-pooling operation and a vertical axis max-pooling operation. These two one-dimensional max-pooling operations are performed by tracking the rank of the values within the window in the RTB-MAXP engine and cascading the maximum operations of the values in CMB-MAXP engine. Both the RBM-MAXP engine and the CMB-MAXP engine were implemented using VHSIC Hardware Description Language (VHDL) and verified by simulations. They have been employed for and tested in our CNN accelerator targeting at the CNN model YOLOv4-CSP-S-Leaky for object detection.

Amid the swift progression of the global automobile industry, the aspiration for lightweight vehicle design has surfaced as a crucial approach to boost fuel efficiency and curtail environmental impact. The objective of this study revolves around achieving holistic optimization of vehicle structure via detailed scrutiny of vehicle process characteristics, amalgamated with a Back Propagation neural network (BPNN) optimized by a genetic algorithm (GA) and a topology optimization (TO) method.The process initiates with an examination of the vehicle's technical characteristics, succeeded by an optimization protocol that couples GA with BPNN. This exhaustive optimization procedure derives the best-fit values of design parameters and identifies the globally optimal lightweight design approach. Subsequently, a comprehensive structural optimization is actualized by modifying the topological distribution of the structure.A simulation experiment of the conceptual car body beam skeleton is conducted as a conclusive step, and the anti-collision capacity of the car body beam skeleton is assessed by measuring deformation intrusion. The findings reveal that in the vehicle topology optimization, a comparative analysis of pre and post-optimization results indicates an improved topological density across diverse sections. The topological density of the vehicle's frontal structure escalates from 0.75 to 0.8. Following optimization, the stiffness assessment records a marginal increase (within the range of 0.02–0.05), implying that the lightweight design does not compromise the general rigidity of the structure.The optimized body structure manifests superior crashworthiness in the crash test: the deformation intrusion diminishes by 27% in front-end collisions. In side impacts, the deformation and intrusion at critical points are reduced by 13.15% to 10% respectively. The comprehensive method proposed herein proves to be an efficient strategy to actualize the lightweight design of vehicles.

Ligand-based virtual screening (LBVS) is a promising approach for rapid and low-cost screening of potentially bioactive molecules in the early stage of drug discovery. Compared with traditional similarity-based machine learning methods, deep learning frameworks for LBVS can more effectively extract high-order molecule structure representations from molecular fingerprints or structures. However, the 3D conformation of a molecule largely influences its bioactivity and physical properties, and has rarely been considered in previous deep learning-based LBVS methods. Moreover, the relative bioactivity benchmark dataset is still lacking. To address these issues, we introduce a novel end-to-end deep learning architecture trained from molecular conformers for LBVS. We first extracted molecule conformers from multiple public molecular bioactivity data and consolidated them into a large-scale bioactivity benchmark dataset, which totally includes millions of endpoints and molecules corresponding to 954 targets. Then, we devised a deep learning-based LBVS called EquiVS to learn molecule representations from conformers for bioactivity prediction. Specifically, graph convolutional network (GCN) and equivariant graph neural network (EGNN) are sequentially stacked to learn high-order molecule-level and conformer-level representations, followed by attention-based deep multiple-instance learning (MIL) to aggregate these representations and then predict the potential bioactivity for the query molecule on a given target. We conducted various experiments to validate the data quality of our benchmark dataset, and confirmed EquiVS achieved better performance compared with 10 traditional machine learning or deep learning-based LBVS methods. Further ablation studies demonstrate the significant contribution of molecular conformation for bioactivity prediction, as well as the reasonability and non-redundancy of deep learning architecture in EquiVS. Finally, a model interpretation case study on CDK2 shows the potential of EquiVS in optimal conformer discovery. The overall study shows that our proposed benchmark dataset and EquiVS method have promising prospects in virtual screening applications.

This paper proposes a radiation pattern synthesis of the almost periodic antenna arrays including mutual coupling effects (that extracted by the Floquet analysis according to our previous work), which principally has a high directivity and large bandwidth. For modeling the given structures, the moment method combined with the Generalized Equivalent Circuit (MoM-GEC) is proposed. The artificial neural network (ANN) as a powerful computational model has been successfully applied to the antenna array pattern synthesis. The results showed that the multilayer feedforward neural networks are rugged and can successfully and efficiently resolve various distinctive complex almost periodic antenna patterns (with different source amplitudes) (in particular, both periodic and randomly aperiodic structures are taken into account). However, the artificial neural network (ANN) is capable of quickly producing the synthesis results using generalization with the early stopping (ES) method. A significant time gain and memory consumption are achieved by using this given method to improve the generalization (called early stopping). To justify this work, several examples are developed and discussed.

Monitoring the integrity of aeronautical structures is fundamental for safety. Structural Health Monitoring Systems (SHMS) perform real-time monitoring functions, but their performances must be carefully assessed. In this work the damage detection performances of a strain-based SHMS, were evaluated on a composite helicopter rotor blade root. The SHMS monitored the bonding between the central core and the surrounding antitorsional layer. A damage detection algorithm was trained through Finite Element analyses performed on the component. The effects of the loads variability and of damages were decoupled by including a load recognition step in the algorithm. Load recognition was accomplished in different ways: version#1 adopted an Artificial Neural Network (ANN) trained on blades in pristine condition; version#2 was trained also on damaged blades; version#3 adopted a calibration matrix method. Anomaly detection, damage assessment, and localization were performed by using an ANN. Results showed a higher load identification and anomaly detection accuracy for version#1 and version#2 compared to version#3. Gaussian noise reduced anomaly detection and damage assessment performances, and reduced damage localization accuracy for bigger damages. The damage detection performance were obtained for a fibre optic based-SHMS, including the trade-off between probability of detection and false alarm rate for different damage sizes.

There are several challenges on how to attain energy efficiency while maintaining balance among factors affecting energy consumption such as power rating and temperature setpoints of cooling units, room temperature and humidity, and indoor air quality (IAQ). A real-time energy and IAQ monitoring system were installed in an educational building to profile the operational power consumption of inverter-based air condition unit (ACU) installed in each room. Polynomial curve and neural fitting regression analysis were applied to the real-time power consumption, indoor temperature and humidity, and carbon dioxide (CO2) level data. The derived models were able to provide the stabilizing indoor thermal and air conditions of the room to reach steady state ACU power consumption. These collected data and calculated parameters can be used to define rules in automating control of cooling appliances for an efficient energy utilization. The regression approaches, using real-time data, have determined the influence of indoor heat conditions and carbon dioxide levels on ACU power consumption. These parameters were utilized to establish consistent values for temperature, humidity, and carbon dioxide levels under stable settings of inverter-based air conditioning units.

Free-space communication is a leading component in global communications. Its advantages relate to a broader signal spread, no wiring, and ease of engagement. Satellite communication services became recently attractive to mega-companies that foresee an excellent opportunity to connect disconnected remote regions, serve emerging machine-to-machine communication, Internet-of-things connectivity, and more. Satellite communication links suffer from arbitrary weather phenomena such as clouds, rain, snow, fog, and dust. In addition, when signals approach the ground station, it has to overcome buildings blocking the direct access to the ground station. Therefore, satellites commonly use redundant signal strength to ensure constant and continuous signal transmission, resulting in excess energy consumption, challenging the limited power capacity generated by solar energy or the fixed amount of fuel. This research proposes LTSM, an artificial recurrent neural network technology that provides a time-dependent prediction of the expected attenuation level due to rain and fog and the signal strength that remained after crossing physical obstacles surrounding the ground station. The satellite transmitter is calibrated accordingly. The satellite outgoing signal strength is based on the predicted signal strength to ensure it will remain strong enough for the ground station to process it. The instant calibration eliminates the excess use of energy resulting in energy savings.

Deep learning classification is the state-of-the-art of machine learning approach. Earlier work proves that the deep convolutional neural network has successfully and brilliantly in different applications such as images or video data. Recognizing and clarifying the remote sensing aspect of the earth's surface and exploit land cover and land use (LCLU). First, this article summarized the remote sensing emerging application and challenges for deep learning methods. Second, we propose four approaches to learn efficient and effective CNNs to transfer image representation on the ImageNet dataset to recognize LCLU datasets. We use VGG16, Inception-ResNet-V2, Inception-V3, and DenseNet201 models to extract features from the EACC dataset. We use pre-trained CNNs on ImageNet to extract features. For feature selection we proposed principal component analysis (PCA) to improve accuracy and speed up the model. We train our model by multi-layer perceptron (MLP) as a classifier. Lastly, we apply the multi-granularity encoding ensemble model. We achieve an overall accuracy of 92.3% for the nine-class classification problem. This work will help remote sensing scientists understand deep learning tools and apply them in large-scale remote sensing challenges

This study was undertaken to forecast waste generation rates of accommodation sector of North Cyprus as a case. Three predictor models including multiple linear regression (MLR), artificial neural networks (ANNs) and central composite design (CCD) were applied to predict the waste generation rate during the lean and peak seasons. ANN showed highest prediction performance, specifically, lowest values of the standard error of prediction (SEP = 2.153), mean absolute error (MAE=1.378) and highest R2 value (0.998) confirmed the accuracy of the model. The analysed wastes were categorised into recyclable, general waste and food residues. The authors estimated the total waste generated during the lean season as 2010.5 kg/day, in which large-sized hotel accounted for largest fraction (66.7%), followed by the medium hotels (19.4%) and guesthouse accounted for smallest part (2.6%). During the peak season, about 49.6% increases in the waste generation rates were obtained. Interestingly, 45% of the wastes were generated by the British tourists while the least waste was generated by African tourists (7.5%). The ANN predicted that the small and large hotels would produce 5.45 and 22.24 tons of waste by the year 2020, respectively. The findings herein are promising and useful in establishing a sustainable waste management system.

Post-training quantization is pivotal in deploying convolutional neural networks for mobile applications. Block-wise reconstruction with adaptive rounding, as employed in prior works like BrecQ and Qdrop, facilitates acceptable 4-bit quantization accuracy. However, adaptive rounding is time-intensive, and its constraint on weight optimization space curtails the potential for quantization performance. The optimality of block-wise reconstruction hinges on the quantization status of subsequent network blocks. In this investigation, we delve into the theoretical underpinnings of the limitations inherent in adaptive rounding and block-wise reconstruction. Our exploration leads to the development of a post-training quantization methodology, designated as AE-Qdrop. This algorithm operates in two distinct phases: block-wise reconstruction and global fine-tuning. The block-wise reconstruction phase introduces a progressive optimization strategy, superseding adaptive rounding, which not only augments quantization precision but also significantly improves quantization efficiency. To mitigate the risk of overfitting, we introduce a random weighted quantized activation mechanism. During the global fine-tuning phase, we account for interdependencies among quantized network blocks. The weight of each network block will be corrected with logit matching and feature matching. Extensive experiments validate that AE-Qdrop achieves high-precision and efficient quantization. For instance, in the case of 2-bit MobileNetV2, AE-Qdrop outperforms Qdrop by achieving a 6.26% enhancement in quantization accuracy and quintupling the quantization efficiency.

Waste heat recovery stands out as a promising technique to tackle both energy shortages and environmental pollution. Currently, this valuable resource, generated from processes like fuel combustion or chemical reactions, is often dissipated into the environment, despite its potential to significantly contribute to the economy. To harness this untapped potential, a travelling-wave thermo-acoustic generator has been designed and subjected to comprehensive experimental analysis. Fifty-two data corresponding to different working conditions of the system were extracted to build ANN, ANFIS and ANN-PSO models. Evaluation of performance metrics reveals that the ANN-PSO model demonstrates the highest predictive accuracy (R^2=0.9959), particularly in relation to output voltage. This research demonstrates the potential of machine learning techniques for the analysis of thermo-acoustic systems. By doing so, it is possible to get an insight into nonlinearities inherent to thermo-acoustic systems. This advancement empowers researchers to forecast the performance characteristics of alternative configurations with a heightened level of precision.

This paper attempts at applying adaptive network-based fuzzy inference system (ANFIS) for analysis of the resonant frequency of a microstrip rectangular patch antenna with two equal size slots which are placed on the patch vertically. The resonant frequency is calculated as the position of slots is shifted to the right and left sides on the patch. As a result , the antenna resonates at more than one frequency . Commonly, machine algorithms based on artificial neural networks are employed to recognize the whole resonant frequencies. However ,they fail to estimate the resonant frequencies correctly as in some cases variations are not very sensible and the resonant frequencies overlap each other . It can be concluded that artificial neural networks could be replaced in such designs by the adaptive network-based fuzzy Inference system due to its high approximation capability and much faster convergence rate.

The correct classification of defects originating from partial discharges (PD) in medium-voltage (MV) switchgears with air insulation (AIS) remains a challenging research topic for scientists worldwide. In this article, the authors simulated four possible defects occurring in the power industry, including one that is a simultaneous combination of two commonly ones. In addition, the correctness of the algorithm was checked by adding a classification class without any fault. The measurement signals were recorded with TEV sensors. The effectiveness of various hy-brid-connected neural networks was tested and discussed: GoogleNet and SqueezeNet based on spectrograms, SAE with FNN, 2D-CNN with LSTM, and hybrid AE combined with CNN and LSTM. The highest effectiveness – approximately 97% – was demonstrated by the GoogleNet and SqueezeNet networks. The research results are expected to form the basis for the development of a universal and wireless capacitive sensor for monitoring the level of PD in switchgears.

Aiming at solving difficulties related to aero-engine classification and identification, the infrared spectra of aero-engine hot jets are measured by telemetry Fourier transform infrared spectrometer, and a six types aero-engine hot jet spectrum data set is generated. The spectrum data set were randomly sampled at a ratio of 8:1:1 to generate training set, validation set and prediction set. In this paper, a Peak Finding Siamese Convolutional Neural Network (PF-SCNN) is designed to match and classify spectrum data. In the first place, the matching pairs of training set and validation set are made independently with the labels of positive and negative samples creating. The spectrum feature extraction and distance similarity calculation are carried out by the Siamese convolutional neural network (SCNN). When entering the prediction phase, the prediction set and training set are integrated to predict in the trained model. The prediction set labels are ultimately determined by matching them with the training set labels having the highest similarities to ensure accurate predictions. To improve the operation efficiency of the SCNN, a peak finding method is introduced to extract the spectrum peaks. Peak positions of high frequency are counted, while the peak data are extracted. The peak data are used to take the place of the original data for training and prediction. The performance measures are defined as Accuracy, Precision, Recall, Confusion matrix and F1-score, and the prediction accuracy was up to 99%. The experimental results indicate that the proposed PF-SCNN is suitable for our spectral dataset and can complete the task of infrared spectrum classification of aero-engine hot jets.

One of the critical tools for early detection and subsequent evaluation of the incidence of lung diseases is chest radiography. This study presents a real-world implementation of a convolutional neural network (CNN) based Carebot Covid app to detect COVID-19 from chest X-ray (CXR) images. Our proposed model takes the form of a simple and intuitive application. Used CNN can be deployed as a STOW-RS prediction endpoint for direct implementation into DICOM viewers. The results of this study show that the deep learning model based on DenseNet and ResNet architecture can detect SARS-CoV-2 from CXR images with precision of 0.981, recall of 0.962 and AP of 0.993.

: The significance of heating load (HL) accurate approximation is the primary motivation of this research to distinguish the most efficient predictive model among several neural-metaheuristic models. The proposed models are through synthesizing multi-layer perceptron network (MLP) with ant lion optimization (ALO), biogeography-based optimization (BBO), dragonfly algorithm (DA), evolutionary strategy (ES), invasive weed optimization (IWO), and league champion optimization (LCA) hybrid algorithms. Each ensemble is optimized in terms of the operating population. Accordingly, the ALO-MLP, BBO-MLP, DA-MLP, ES-MLP, IWO-MLP, and LCA-MLP presented their best performance for population sizes of 350, 400, 200, 500, 50, and 300, respectively. The comparison was carried out by implementing a ranking system. Based on the obtained overall scores (OSs), the BBO (OS = 36) featured as the most capable optimization technique, followed by ALO (OS = 27) and ES (OS = 20). Due to the efficient performance of these algorithms, the corresponding MLPs can be promising substitutes for traditional methods used for HL analysis.

Scoring targets in shooting sports is a crucial and time-consuming task that relies on manually counting bullet holes. This paper introduced an automatic score detection model using object detection techniques. The performance of seven models belonging to two different architectural setups was compared. Models like YOLOv8n, YOLOv8s, YOLOv8m, RetinaNet-50, and RetinaNet-101 are single-shot detectors, while Faster RCNN-50 and Faster RCNN-101 belong to the two-shot detectors category. The dataset was manually captured from the shooting range and expanded by generating more versatile data using Python code. Before the dataset was trained to develop models, it was resized (640x640) and augmented using Roboflow API. The trained models were then assessed on the test dataset, and their performance was compared using matrices like mAP50, mAP50-90, precision, and recall. The results showed that YOLOv8 models can detect multiple objects with good confidence scores. Among all, YOLOv8m performed the best with the highest mAP50 value of 96.7%, followed by the performance of YOLOv8s with the mAP50 value of 96.5%. It is suggested that if the system is to be implemented in a real-time environment, YOLOv8s is a better choice since it took significantly less inference time (2.3ms) than YOLOv8m (5.7ms) and yet generated the competitive mAP50.

The Internet of Things (IoT) aims to connect all objects and facilitate data interchange, with data accuracy being paramount. While IoT has seen extensive adoption in conventional sectors, its application in the petroleum industry, especially with downhole equipment, poses challenges due to the extreme high-temperature and high-pressure conditions, making accurate measurements elusive. This Paper we aim to enhance the accuracy of downhole pressure measurements in slim-hole drilling parameter monitoring, mitigate the temperature influences on strain gauges, and surmount the intrinsic limitations of the traditional BP neural network, such as sluggish convergence speeds, high sensitivity to initial weights and biases, unstable learning rates, and a tendency to fall into local minima. This study introduces a temperature compensation model using a chaotic-initiated adaptive whale optimization algorithm (C-I-WOA) to optimize a BP neural network, which is henceforth referred to as the C-I-WOA-BP model. Initially, we harnessed chaotic mapping techniques to refine the initialization process of the whale optimization algorithm (WOA) population, thereby bolstering the heterogeneity of the algorithm. Subsequently, we integrated an adaptive weight recalibration mechanism, refined the encompassing methodology of the WOA, and bolstered its capabilities for expansive global scouting and meticulous local refinement in the solution arena. Subsequently, the WOA was used to optimize the weight coefficients of the BP neural network. Finally, we conducted a detailed comparative analysis by juxtaposing the performance metrics of the BP, C-WOA-BP, I-WOA-BP, and C-I-WOA-BP networks. The experimental results indicate that the C-I-WOA-BP model demonstrates faster convergence speed, superior global search capabilities, and enhanced local exploitation potential compared to the other models. Additionally, it provides a more stable and accurate reduction in temperature-induced errors in drilling pressure parameter measurements.

Despite numerous studies in statistical downscaling methodology, there remains a lack of methods that can downscale from precipitation modeled in global climate models to regional level high resolution gridded precipitation. This paper reports a novel downscaling method using a Generative Adversarial Network (GAN), CliGAN, which can downscale large-scale annual maximum precipitation given by simulation of multiple atmosphere-ocean global climate models (AOGCM) from Coupled Model Inter-comparison Project 6 (CMIP6) to regional-level gridded annual maximum precipitation data. This framework utilizes a convolution encoder-dense decoder network to create a generative network and a similar network to create a critic network. The model is trained using an adversarial training approach. The critic uses the Wasserstein distance loss function and the generator is trained using a combination of adversarial loss Wasserstein distance, structural loss with the multi-scale structural similarity index (MSSIM), and content loss with the Nash-Sutcliff Model Efficiency (NS). The MSSIM index allowed us to gain insight into the model’s regional characteristics and shows that relying exclusively on point-based error functions, widely used in statistical downscaling, may not be enough to reliably simulate regional precipitation characteristics. Further use of structural loss functions within CNN-based downscaling methods may lead to higher quality downscaled climate model products.

This article explores the crucial task of estimating State of Charge (SOC) in Hybrid Electric Vehicles (HEVs) and examines the applicability of various regression models for this purpose. We delve into the strengths and limitations of Linear Regression, Support Vector Regression (SVR), and Neural Network Regression (NNR) in the context of SOC estimation. Linear regression provides a simple and interpretable baseline, SVR extends this to nonlinear relationships, while NNR emerges as a powerful tool with adaptive capabilities. The choice of model depends on factors such as data characteristics, interpretability, and computational resources. As the field evolves, the article advocates for a nuanced approach, possibly incorporating hybrid models, to achieve robust and accurate SOC estimation, contributing to the ongoing enhancement of HEV efficiency and sustainability.

Computer vision models of salient object detection attempt to mimic the ability of the human visual system to select relevant objects in images. To this end, the development of deep neural networks on high-end computers has recently made it possible to achieve high performance. However, it remains a challenge to develop deep neural network models of the same performance for devices with much more limited resources. In this work, we propose a new approach for a lightweight salient object detection neural network model, inspired by the cone and spatial opponent processes of the primary visual cortex (V1), that inextricably link color and shape in human color perception. Our proposed model, namely CoSOV1net, is trained from scratch, without using backbones from image classification or other tasks. Experiments, on the most widely used and challenging datasets for salient object detection, show that CoSOV1Net achieves competitive performance (i.e. Fβ=0.931 on the ECSSD dataset) with state-of-the-art salient object detection models, while having low number of parameters (1.14M), low FLOPS (1.4G) and high FPS (211.2) on GPU (nvidia Geforce RTX 3090 TI) compared to the state-of-the-art in the salient object detection or lightweight salient object detection task. Thus, CoSOV1net turns out to be a lightweight salient object detection that can be adapted to mobile environments and resource-constrained devices.

This study aims to develop a hybrid approach based on backpropagation Artificial Neural Network (ANN) and spatial analysis techniques to predict particulate matter of size 2.5 µm (PM2.5) from vehicle exhaust emissions in the State of California using Aerosol Optical Depth (AOD) and several climatic indicators (relative humidity, temperature, precipitation, and wind speed). The PM2.5 data were generated using Motor Vehicle Emission Simulator (MOVES), the measured climatic variables and AOD were obtained from the California Irrigation Management Information System (CIMIS), and NASA’s Moderate Resolution Spectroradiometer (MODIS). The data were resampled to a seasonal format and downscaled over grids of 10 by 10 to 150 by 150, and precipitation was determined to be the most important independent variable. Coefficient of determination ( ), Mean Absolute Percentage Error (MAPE), and Root Mean Square Error (RMSE) were used to assess the quality of the ANN prediction model. The model peaked at winter seasons with = 0.984, RMSE = 0.027, and MAPE = 25.311, whereas it had the lowest performance in summer with = 0.920, RMSE = 0.057, and MAPE = 65.214. These results indicate that the ANN model can accurately predict the PM2.5 concentration and can be used to forecast future trends.

Electricity demand prediction is vital for energy production management and proper exploitation of the present resources. Recently, several novel machine learning (ML) models have been employed for electricity demand prediction to estimate the future prospects of the energy requirements. The main objective of this study is to review the various ML models applied for electricity demand prediction. Through a novel search and taxonomy, the most relevant original research articles in the field are identified and further classified according to the ML modeling technique, perdition type, and the application area. A comprehensive review of the literature identifies the major ML models, their applications and a discussion on the evaluation of their performance. This paper further makes a discussion on the trend and the performance of the ML models. As the result, this research reports an outstanding rise in the accuracy, robustness, precision and the generalization ability of the prediction models using the hybrid and ensemble ML algorithms.

Machine learning (ML) has a large capacity to learn and analyze a large volume of data. This study aimed to train different algorithms to discriminate between healthy and pathologic corneal images by evaluating digitally processed spectral-domain optical coherence tomography (SD-OCT) corneal images. A set of 22 SD-OCT images belonging to a random set of corneal pathologies was compared to 71 healthy corneas (control group). A binary classification method was applied; three approaches of ML were used. Once all images were analyzed, representative areas from every digital image were also processed and analyzed for a statistical feature comparison between healthy and pathologic corneas. The best performance was obtained from transfer learning - support vector machine (TL-SVM) (AUROC = 0.94, SPE 88%, SEN 100%) and transfer learning – random forest (TL- RF) method (AUROC = 0.92, SPE 84%, SEN 100%), followed by convolutional neural network (CNN) (AUROC = 0.84, SPE 77%, SEN 91%) and random forest (AUROC = 0.77, SPE 60%, SEN 95%). The highest diagnostic accuracy in classifying corneal images was achieved with the TL-SVM and the TL-RF models. In image classification, CNN was a strong predictor. This pilot experimental study developed a systematic mechanized system to discern pathologic from healthy corneas.

For analysis tasks, time counts are of interest — values recorded at some, usually equidistant, points in time. The calculation can be performed at various intervals: after a minute, an hour, a day, a week, a month, or a year, depending on how much detail the process should be analyzed. In time series analysis problems, we deal with discrete-time, when each observation of a parameter forms a time frame. The same can be said about the behavior of Covid-19 over time.In this paper, we solve the problem of predicting Covid-19 diseases in the world using neural networks. This approach is useful when it is necessary to overcome difficulties related to non-stationarity, incompleteness, unknown distribution of data, or when statistical methods are not completely satisfactory. The problem of forecasting is solved with the help of the analytical platform Deductor Studio, developed by specialists of the company Intersoft Lab of the Russian Federation. When solving this problem, appropriate methods were used to clean the data from noise and anomalies, which ensured the quality of building a predictive model and obtaining forecast values for tens of days ahead. The principle of time series forecasting was also demonstrated: import, seasonal detection, cleaning, smoothing, building a predictive model, and predicting Covid-19 diseases in the world using neural technologies for 30 days ahead.

Delay in data transmission is one of key performance indicators (KPIs) of a network. The planning and project value of delay in network management is of crucial importance for the optimal allocation of network resources and their performance focuses. To create optimal solutions, predictive models, which are currently most often based on machine learning (ML), are used. This paper aims to investigate the training, testing and selection of the best predictive delay model for a VoIP service in an Long Term Evolution (LTE) network using three ML techniques - Neural Networks (NN), Support Vector Machines (SVM) and k-Nearest Neighbors (k-NN). The space of model input variables is optimized by dimensionality reduction techniques: RReliefF algorithm, Backward selection via the recursive feature elimination algorithm and the Pareto 80/20 rule. A three-segment road in the geo-space between the cities of Banja Luka (BL) and Doboj (Db) in the Republic of Srpska (RS), Bosnia and Herzegovina (BiH), covered by the cellular network (LTE) of the M:tel BL operator was chosen for the case study. The results show that, in all three optimization approaches, the k-NN model is selected as the best solution. For the RReliefF optimization algorithm, the best model has 6 inputs and minimum relative error (RE), RE=0.109; for the Backward selection via the recursive feature elimination algorithm, the best model has 4 inputs and RE=0.041; and for the Pareto 80/20 rule, the best model has 11 inputs and RE= 0.049. The comparative analysis of the results concludes that according to observed criteria for the selection of the final model, the best solution is an approach to optimizing the number of predictors based on the Backward selection via the recursive feature elimination algorithm.

In the remote sensing domain, it is crucial to complete semantic segmentation on the raster images, e.g., river, building, forest, etc, on raster images. A deep convolutional encoder--decoder (DCED) network is the state-of-the-art semantic segmentation method for remotely sensed images. However, the accuracy is still limited, since the network is not designed for remotely sensed images and the training data in this domain is deficient. In this paper, we aim to propose a novel CNN for semantic segmentation particularly for remote sensing corpora with three main contributions. First, we propose applying a recent CNN called a global convolutional network (GCN), since it can capture different resolutions by extracting multi-scale features from different stages of the network. Additionally, we further enhance the network by improving its backbone using larger numbers of layers, which is suitable for medium resolution remotely sensed images. Second, "channel attention'' is presented in our network in order to select the most discriminative filters (features). Third, "domain-specific transfer learning'' is introduced to alleviate the scarcity issue by utilizing other remotely sensed corpora with different resolutions as pre-trained data. The experiment was then conducted on two given datasets: (i) medium resolution data collected from Landsat-8 satellite and (ii) very high resolution data called the ISPRS Vaihingen Challenge Dataset. The results show that our networks outperformed DCED in terms of $F1$ for 17.48% and 2.49% on medium and very high resolution corpora, respectively.

This paper focused on the finite element analysis of structural system in extreme loading condition. Two different stud shape and thicknesses were analyzed under blast. The stud thickness such as 1.19 mm and 1.5 mm were modelled and analyzed using ABAQUS 6.14. FEM is a tool which predicts the engineering physics of the real structure. To validate the finite element modelling performed by authors, a reference work published by earlier researchers on cold-formed steel stud wall is considered and examined in the present study. The novelty of this study was web corrugation and influence of flange width on stud. The models mimic like an air bag in a car to delay the pressure timing inside the stud wall. The mass of explosive used as 1.56 kg at a standard scaled distance. Time versus displacement was captured out at four locations in the stud wall. One of the objectives is to develop mathematical model to validate the deformation of stud under blast loading. Two mathematical models were validated using Artificial Neural Network (ANN). The results captured in ANN model was error histogram, regression plot, best performance fit and training data. The models were capable of resisting the moderate blast load. The response surface methodology (RSM) was employed to evaluate model performance Regression equations are useful for predicting future trends and outcomes, which is crucial for planning and decision-making. The primary goal of this work is to evaluate cold-formed steel stud walls with varying stud sizes under blast loading using finite element analysis and validated by ANN and RSM.

Byers Peninsula (62°34’-62°40’S-60°54’61°13’W), 60 km2 in size, is considered one of the largest ice-free areas in Antarctica. Since 2006, the Spanish Polar Program has taken part in a large number of environmental studies involving effects of climate changes, limnology and microbiology, live cycles, but not albedo. Surface albedo is one of the key physical parameters in the surface energy budget of polar regions. Most of Antarctica is covered by ice sheets; only about 0.44% of the area is permanently ice-free. However, in maritime Antarctica, the ice-free areas, corresponding to small islands, peninsulas, and coastal beaches, account for about 3% of the territory. To incorporate the contribution of these areas into global albedo models, it is necessary to relate the soil properties of the ice-free areas to the surface albedo response. Also, mapping soil properties and albedo have and special interest in these ice-free areas. Image classification using machine learning methods trained with georeferenced soil data could be useful for mapping soil properties and albedo in multispectral optical satellite images. A shallow neural network implemented using the Keras Python module was used to define and train models of soil properties using 15 explanatory variables corresponding to bands and spectral indices of a Sentinel image and a population of 49 soil samples taken from the top 5 centimeters of the soil profile. The soil samples were analyzed in the laboratory and a spectral library in the Vis-Nir range (350-2500 nm) was created. The albedo of the samples was integrated from the ADS spectra. At the same time, a linear regression model of albedo using the soil properties as explanatory variables was performed. The R2 fit of this new model was about 0.82 and the error of estimation was 4.1. The model was extended to the entire Byers Peninsula using the ML soil property models as explanatory variables. The RMSE of the extended models increased up to 8.2. We think that the main cause of this increase (2.6 points) is related to the error propagation, phenomenon due to the use of models as explanatory variables. However, another important part of the error increase could be due, among other reasons, to the use of an image that, although corrected, is not completely free of clouds.

This work addresses the challenge of EEG visual multiclass classification into 40 classes for Brain-Computer interface applications, by using deep learning architectures. The visual multiclass classification approach offers BCI applications a significant advantage, since it allows the supervision of more than one BCI interaction, considering that each class label supervises a BCI task. However, because of the nonlinearity and nonstationarity of EEG signals, performing multiclassifiers based on EEG features remains a significant challenge for BCI systems. In the present work, mutual information-based discriminant channel selection and Minimum-Norm Estimate algorithms are implemented to select discriminant channels and enhance EEG data. Hence, deep EEGNet and Convolutional-recurrent neural networks are implemented separately to classify EEG data of image visualization into 40 labels. By using the k-fold cross-validation approach, average classification accuracies of 94.8% and 89.8% were obtained by implementing the aforementioned network architectures, respectively. Satisfactory results obtained with this method offer a new implementation opportunity for multi-task BCI applications by utilizing a reduced number of channels (<50%), compared to those presented in the related literature where the whole set of channels is used.

Light Detection and Ranging (LiDAR) technology, a cutting-edge advancement in mobile applications, presents a myriad of compelling use cases, including enhancing low-light photography, capturing and sharing 3D images of fascinating objects, and elevating the overall augmented reality (AR) experience. However, its widespread adoption has been hindered by the prohibitive costs and substantial power consumption associated with its implementation. To surmount these obstacles, this paper proposes a low-power, low-cost, SPAD-based system-on-chip (SoC) which packages the microlens arrays (MLA) and incorporates with a light-weight RGB-guided sparse depth imaging completion neural network for 3D LiDAR imaging. The proposed SoC integrates an 8x8 Single-Photon Avalanche Detectors (SPADs) macro pixel array with time-to-digital converters (TDC) and charge pump, fabricated using a 180nm bipolar-CMOS-DMOS (BCD) process. A random MLA-based homogenizing diffuser efficiently transforms Gaussian beams into flat-topped beams with a 45° field of view (FOV), enabling flash projection at the transmitter. To further enhance resolution and broaden application possibilities, a lightweight neural network employing RGB-guided sparse depth complementation is proposed, enabling a substantial expansion of image resolution from 8x8 to quarter video graphics array level (QVGA; 256x256). Experimental results demonstrate the effectiveness and stability of the hardware encompassing the SoC and optical system, as well as the lightweight features and accuracy of the algorithmic neural network. This integrated state-of-the-art hardware-software solution offers a promising and inspiring foundation for developing consumer-level 3D imaging applications.

In recent times, with the increase of Artificial Neural Network (ANN), deep learning has brought a dramatic twist in the field of machine learning by making it more Artificial Intelligence (AI). Deep learning is used remarkably used in vast ranges of fields because of its diverse range of applications such as surveillance, health, medicine, sports, robotics, drones etc. In deep learning, Convolutional Neural Network (CNN) is at the center of spectacular advances that mixes Artificial Neural Network (ANN) and up to date deep learning strategies. It has been used broadly in pattern recognition, sentence classification, speech recognition, face recognition, text categorization, document analysis, scene, and handwritten digit recognition. The goal of this paper is to observe the variation of accuracies of CNN to classify handwritten digits using various numbers of hidden layer and epochs and to make the comparison between the accuracies. For this performance evaluation of CNN, we performed our experiment using Modified National Institute of Standards and Technology (MNIST) dataset. Further, the network is trained using stochastic gradient descent and the backpropagation algorithm.

Keywords: Аrtificial neural network, electric drive system, database, stability, training algorithm, technological mechanisms.

In this article, a method for the moisture mapping of the soil surface of agrophytocenosis was proposed using neural network based on synchronized radar and multispectral optoelectronic data of Sentinel-1,2. To verify the developed method, data from two experimental plots were used. These plots were located two on irrigated soybean crops. The first of them was located on the right bank (1st plot) and the second one on the left bank (2nd plot) of the down part of Volga River. Two experimental soil moisture geo-datasets were done by measurements and geo-referencing points using gravimetric method (1st plot) and with proximal sensing method (2nd plot) using Soil Moisture Sensor ML3-KIT (THETAKIT, Delta). The soil moisture retrieval algorithm was based on the use of a neural network to predict reflection coefficient of an electromagnetic wave from the soil surface, followed by inversion into soil moisture using a dielectric model that takes into account the soil texture. The input parameter of the neural network was the ratio of the microwave radar vegetation index (calculated on the basis of Sentinel-1 data) to the index (calculated on the basis of data of multispectral optoelectronic channels 8 and 11 of Sentinel-2). Such way calculated index reveals showed a significantly greater dependence on soil moisture than on vegetation height that was been used in previous studies. The retrieved values of soil moisture were compared with the soil moisture measured in-situ. The proposed method with a determination coefficient of 0.44-0.65 and a standard deviation of 2.4%-4.2% for the 1st plot as well as with and of the same metrics for the 2nd allows predicting the soil moisture of both a test plots covered by soybean plants, relative to soil moisture measured in-situ. The conducted research created the scientific basis for a new technology for remote sensing the moisture content of the soil surface of agrophytocenosis as an element of the precision farming system and agroecology.

Deep Learning (DL) oriented document processing is widely used in different fields for extraction, recognition, and classification processes from raw corpora of data. The article examines the application of deep learning approaches, based on different neural network methods, including Gated Recurrent Unit (GRU), Long Short-Term Memory (LSTM), and Convolutional Neural Networks (CNN). The compared models were combined with two different word embedding techniques, namely: Bidirectional Encoder Representations from Transformers BERT and Gensim Word2Vec. The models are designed to evaluate the performance of architectures based on neural network techniques for the classification of CVs of Moroccan engineering students at ENSAK(National School of Applied Sciences of Kenitra, Ibn Tofail University). The used dataset included resumes collected from engineering students at ENSAK in 2023 for a project on the employability of Moroccan engineers in which new approaches were applied, especially machine learning, deep learning, and big data. Accordingly, 867 resumes were collected from five specialties of study (Electrical Engineering, Networks and Systems Telecommunications, Computer Engineering, Automotive Mechatronics Engineering, Industrial Engineering). The results revealed good performance of the proposed models based on the BERT embedding approach compared to models based on the Gensim Word2Vec embedding approach. Accordingly, the CNN-GRU/Bert model achieved slightly better accuracy with 0.9251 compared to other hybrid models.

Recent studies have proven that synthetic aperture radar (SAR) automatic target recognition (ATR) models based on deep neural networks (DNN) are vulnerable to adversarial examples. However, existing attacks are easily failed in the case where adversarial perturbations cannot be fully fed to victim models. We call this situation perturbation offset. Moreover, since background clutter takes up most of the areas in SAR images and has low relevance to recognition results, fooling models with global perturbations is quite inefficient. This paper proposes a semi-whitebox attack network, called Universal Local Adversarial Network (ULAN), to generate universal adversarial perturbations (UAP) for the target regions of SAR images. In the proposed network, we calculate the model’s attention heatmaps through layer-wise relevance propagation (LRP), which is used to locate the target regions of SAR images that have high relevance to recognition results. In particular, we utilize a generator based on the U-Net to learn the mapping from noise to UAPs and craft adversarial examples by adding the generated local perturbations to target regions. Experiments indicate that the proposed method fundamentally prevents perturbation offset and achieves comparable attack performance to conventional global UAPs by perturbing only a quarter or less of SAR image areas.

This article analyzes and compares the integration of two different maximum power point tracking (MPPT) control methods, which are tested under partial shading and fast ramp conditions. These MPPT methods are designed by Improved Particle Swarm Optimization (IPSO) and a combination technique between Neural Network and Perturb & Observe method (NN_P&O). These two methods are implemented and simulated for photovoltaic systems (PV), where various system responses, such as: voltage and power are obtained. The MPPT techniques were simulated using Matlab/Simulink environment. A comparison of the performance of IPSO and NN_P&O algorithms is carried out to confirm the best accomplishment of the two methods in terms of speed, accuracy and simplicity.

The detection performance of small objects in remote sensing images has not been satisfactory compared to large objects, especially in low-resolution and noisy images. A generative adversarial network (GAN)-based model called enhanced super-resolution GAN (ESRGAN) showed remarkable image enhancement performance, but reconstructed images usually miss high-frequency edge information. Therefore, object detection performance showed degradation for small objects on recovered noisy and low-resolution remote sensing images. Inspired by the success of edge enhanced GAN (EEGAN) and ESRGAN, we applied a new edge-enhanced super-resolution GAN (EESRGAN) to improve the quality of remote sensing images and used different detector networks in an end-to-end manner where detector loss was backpropagated into the EESRGAN to improve the detection performance. We proposed an architecture with three components: ESRGAN, EEN, and Detection network. We used residual-in-residual dense blocks (RRDB) for both the ESRGAN and EEN, and for the detector network, we used a faster region-based convolutional network (FRCNN) (two-stage detector) and a single-shot multibox detector (SSD) (one stage detector). Extensive experiments on a public (car overhead with context) dataset and another self-assembled (oil and gas storage tank) satellite dataset showed superior performance of our method compared to the standalone state-of-the-art object detectors.