In this work, we propose a Fog-enabled UAV-as-a-Service (FU-Serve) architecture to address the issues of data transmission delay for serving time-critical Internet of Things (IoT) applications. Traditionally, in a UAV-as-a-Service (UaaS) platform, different UAVs host heterogeneous sensors, which sense the physical phenomenon and transmit the sensed data to a centralized entity. Transmission of data from the sensors to the centralized entity and making any decision for an application consumes a significant amount of time. Consequently, the traditional UaaS architecture is unsuitable for serving time-critical IoT applications such as transportation, healthcare, and industries. To address these issues of service latency for time-critical IoT applications, we present the FU-Serve architecture by introducing the concept of fog computing in the UaaS platform. We discuss all the components of FU-Serve elaborately in this paper. Additionally, we architect optimal and dynamic fog node selection mechanisms for FU-Serve, which reduce the transmission delay in the networks. The simulation results show that the FU-Serve outperforms by $75\%$ compared to the traditional UaaS platform.

The decade of research concentrating on the area of Easter Slavonia revealed abundance of large on complex late Neolithic sites. It changed profoundly how we perceive late Neolithic settlement and space size and organization. As s first step in particular site detection we mostly conducted aerial reconnaissance and satellite image analysis but also data from the field such as excavation. On detected potential sites we conducted intensive field surveys and confirmed the attribution to the late Neolithic by surface finds. On those confirmed sites in the vicinity of Đakovo, Croatia, magnetic survey was conducted on 5 sites so far and the results confirmed presence of large scale late Neolithic settlements with complex spatial organization and enclosure(s). Most complex remains so far is site Gorjani Kremenjača and Topole which we present in this paper where a remains of one or two settlements covers the area of 70 hectares and is currently in the process of formal protection as a cultural landscape by the Ministry of Culture and Media of the Republic of Croatia. Special focus of the paper is the application of remote sensing in detection, archaeological confirmation and protection of the position Gorjani Topole.

Multi-Object Tracking (MOT) is a key technology for Unmanned Aerial Vehicles (UAVs). Traditional tracking-by-detection methods firstly employ an object detector to retrieve targets in each image and then track them based on a matching algorithm. Recently, the popular multi-task learning methods has been dominating this area since they can detect targets and extract Re-Identification (Re-ID) features in a computationally efficient way. However, the detection task and the tracking task have conflicting requirements on image features, leading to the poor performance of the joint learning model compared to separate detection and tracking methods. The problem is more severe when it comes to UAV images due to the presence of irregular motion of large number of small targets. In this paper, we propose a balanced Joint Detection and Re-ID learning (JDR) network to address the MOT problem in UAV vision. To better handle the non-uniform motion of objects in UAV videos, the Set-Member Filter is applied which describes object state as a bounded set. An appearance matching cascade is then proposed based on target state set. Furthermore, a Motion-Mutation module is designed to address the challenges posed by the abrupt motion of UAV. Extensive experiments on the VisDrone-MOT2019 dataset demonstrate that our proposed model, termed as SMFMOT, outperforms state-of-the-arts by a large margin and achieves superior performance on the MOT tasks in UAV videos.

Unmanned aerial vehicles (UAV) have emerged as a solution to day to day survey tasks, allowing users to visualize phenomena in real-time. This paper explores the capabilities of UAV or drones in the collection of accurate, geo-tagged data quickly, including photogrammetry software processes to deliver standardized data output. In order to explore the capabilities of UAV, Gatu Township in Centenary, Muzabani District of Zimbabwe was chosen from the national mapping topographic series. This study demonstrates the efficiency of data collection using drones, generate 2D orthomosaics in real time, so that analysts can easily visualize land cover and identify any changes, map and model large areas to produce data for 2D and 3D models. The recent development of innovative optical image processing has further lowered the costs high resolution topographic surveys.

The coombs (in Catalan comes) form an integral part of the landscape. These are small valleys, with the deepest and flattest part cultivated for agriculture. They are predominantly found in the most arid regions of Catalonia, particularly around the cities of Tarragona and Lleida. They are generally dry land areas. We propose that these coombs gained significance in the early centuries of the Middle Ages, a period marked by intense erosion. It is important to identify their locations and establish connections with medieval written documents, the settlements that thrived throughout the Middle Ages, archaeological sites such as necropolises, and the valuable data derived from to-ponymy studies. We think that understanding these coombs can greatly contribute to under-standing how the territory was organized in the Middle Ages, the extent of cultivated lands that were established, and the distribution of many settlements. Ultimately, this knowledge can aid in acquiring better understanding of the current landscape.

Deep learning techniques are increasingly being recognized as effective image classifiers. Aside from their successful performance in past studies, the accuracies have varied in complex environments in comparison with the popularly applied machine learning classifiers. This study seeks to explore the feasibility for using a U-Net deep learning architecture to classify bi-temporal high resolution county scale aerial images to determine the spatial extent and changes of land cover classes that directly or indirectly impact tidal marsh. The image set used in the analysis is a collection of a 1-m resolution collection of National Agriculture Imagery Program (NAIP) tiles from 2009 and 2019 covering Beaufort County, South Carolina. The U-net CNN classification results were compared with two machine learning classifiers, the Random Trees (RT) and the Support Vector Machine (SVM). The results revealed a significant accuracy advantage in using the U-Net classifier (92.4%) as opposed to the SVM (81.6%) and RT (75.7%) classifiers for overall accuracy. From the perspective of a GIS analyst or coastal manager, the U-Net classifier is now an easily accessible nad powerful tool for mapping large areas. Change detection analysis indicated little areal change on marsh extent, though increased land development throughout the county has the potential to negatively impact the health of the marshes. Future work should explore applying the constructed U-Net classifier to coastal environments in large geographic areas, while also implementing other data sources (e.g., LIDAR, multispectral data) to enhance classification accuracy.

CNN-based methods have dominated the field of aerial scene classification for the past few years. While achieving remarkable success, CNN-based methods suffer from excessive parameters and notoriously rely on large amounts of training data. In this work, we introduce few-shot learning to the aerial scene classification problem. Few-shot learning aims to learn a model on base-set that can quickly adapt to unseen categories in novel-set, using only a few labeled samples. To this end, we proposed a meta-learning method for few-shot classification of aerial scene images. First, we train a feature extractor on all base categories to learn a representation of inputs. Then in the meta-training stage, the classifier is optimized in the metric space by cosine distance with a learnable scale parameter. At last, in the meta-testing stage, the query sample in the unseen category is predicted by the adapted classifier given a few support samples. We conduct extensive experiments on two challenging datasets: NWPU-RESISC45 and RSD46-WHU. The experimental results show that our method yields state-of-the-art performance. Furthermore, several ablation experiments are conducted to investigate the effects of dataset scale, the impact of different metrics and the number of support shots; the experiment results confirm that our model is specifically effective in few-shot settings.

Based on the theory of turbulence, equations are derived for estimations of turbulent fluctuations of the longitudinal and transverse components of the wind velocity during ideal hovering of a quadcopter in a turbulent atmosphere. We present the results of experiments which were carried out on the territory of the Geophysical Observatory of Institute of Monitoring of Climatic and Ecological Systems, Siberian Branch, Russian Academy of Sciences, located in Tomsk Akademgorodok on the territory with complex orography, in a parkland zone with buildings of research institutes and motorways. Time series of turbulent fluctuations of the longitudinal and transverse components of wind velocity fluctuations were received with the use of an automated weather station, and time series of estimates of these components, from data of a DJI Phantom 4 Pro quadcopter during hovering. According to the automated weather station data, anisotropy was observed in one experiment during measurements in the atmosphere, but this phenomenon was not observed in the other experiment: the fluctuation spectra of all components of wind speed fluctuations coincide. The spectra of fluctuations of the longitudinal and transverse wind velocity components based on the automated weather station data and UAV telemetry are presented. The fluctuation spectra of these components for the automated weather station data and quadcopter generally coincide. The behavior of the spectra coincides with the spectrum which corresponds to Kolmogorov–Obukhov “–5/3” law within the inertial range. The turbulent spectra of the wind velocity fluctuations obtained with the use of the automatic weather station and with the unmanned aerial vehicle differ in the high-frequency spectral region.

Terrestrial laser scanning (TLS) provides detailed three-dimensional representation of the surrounding forest structure. However, due to close-range hemispherical scanning geometry, the ability of TLS technique to comprehensively characterize all trees and especially the upper parts of forest canopy is often limited. In this study, we investigated how much forest characterization capacity can be improved in managed Scots pine (Pinus sylvestris L.) stands if TLS point cloud is complemented with a photogrammetric point cloud acquired from above the canopy using unmanned aerial vehicle (UAV). In this multisensorial (TLS+UAV) close-range sensing approach, the used UAV point cloud data was considered feasible especially in characterizing the vertical forest structure and improvements were obtained in estimation accuracy of tree height as well as plot-level basal-area weighted mean height (H_g) and mean stem volume (V_mean). Most notably the root mean square error (RMSE) in H_g improved from 0.88 m to 0.58 m and the bias improved from -0.75 m to -0.45 m with the multisensorial close-range sensing approach. However, in managed Scots pine stands the mere TLS captured also the upper parts of the forest canopy rather well. Both approaches were capable of deriving stem number, basal area, V_mean, H_g and basal area-weighted mean diameter with a relative RMSE less than 5.5% for all of the sample plots. Although the multisensorial close-range sensing approach mainly enhanced characterization of forest vertical structure in single-species, single-layer forest conditions, representation of more complex forest structures may benefit more from point clouds collected with sensors of different measurement geometries.

Unmanned Aerial Vehicle (UAV) based remote sensing (RS) studies in glaciology are mainly focusing on obtaining accurate high-resolution data from UAV images. Studies for identifying and minimising the challenges faced during the UAV-based RS data acquisition survey on inaccessible and harsh terrains of mountain glaciers is limited. This study aims to examine the practical challenges faced during UAV surveys of glaciers and derive strategies to minimize them. To the authors' knowledge, this is the first study that addresses such problems over the Himalayan region. Here, the UAV surveys were conducted using a fixed-wing commercial-grade off-the-shelf UAV (eBee plus, SenseFly) on three glacier sites (East Rathong, Hamtah and Panchinala-A) located in different zones and climate regimes lying within the Indian part of Himalayas. From UAV collected photos, the study was able to generate ultra-high-resolution ortho-mosaicked images and Digital Elevation Models (DEMs) at 0.1m GSD. UAV-derived DEMs was able to achieve vertical (horizontal) accuracy of 0.45 and 0.21 m (0.15 and 0.1 m) with 3 and 6 ground control points (GCPs) for an area of 0.75 km2 and 1.38 km2. Accuracy assessment of UAV DEMs generated with and without GCPs indicate that GCPs are must to obtain decimetre level accurate DEM especially on glaciers with steep-valleyed terrains. The utility of the obtained ultra-high-resolution ortho-mosaicked images was demonstrated by generating glacier surface feature maps. Based on the challenges observed during UAV surveys, the study identifies and recommends best-suited locations on a glacier and its adjacent regions for conducting UAV surveys efficiently in the glaciated terrain of Himalayas and possibly beyond. Recommendations reported in this article shall minimise the challenges faced and involved risks for data acquisition and thus enable UAVs to cover more glaciated area successfully.

UAVNs (unmanned aerial vehicle networks) may become vulnerable to threats and attacks due to their characteristic features such as high mobility, highly dynamic network topology, and open-air wireless environments. Since previous work has focused on classical and metaheuristic-based approaches, none of these approaches have a self-adaptive approach. In this article, we examine the challenges of cyber detection methods to secure UAVNs and review exiting security schemes proposed in the current literature. Furthermore, we propose an agent-based self-protective method (ASP-UAVN) for UAVNs that is based on the Human Immune System (HIS). In ASP-UAS, the safest route from the source UAV to the destination UAV is chosen according to a self-protective system. In this method, a multi-agent system using an Artificial Immune System (AIS) is employed to detect the attacking UAV and choose the safest route. In the proposed ASP-UAVN, the route request packet (RREQ) is initially transmitted from the source UAV to the destination UAV to detect the existing routes. Then, once the route reply packet (RREP) is received, a self-protective method using agents and the knowledge base is employed to choose the safest route and detect the attacking UAVs. The method is evaluated here via extensive simulations carried out in the NS-3 environment. The experimental results of four scenarios demonstrated that the ASP-UAS increases the Packet Delivery Rate (PDR) by more than 17.4, 20.8, and 25.91%, and detection rate by more than 17.2, 23.1, and 29.3%, and decreases the Packet Loss Rate (PLR) by more than 14.4, 16.8, and 20.21%, the false-positive and false-negative rate by more than 16.5, 25.3, and 31.21% those of SUAS-HIS, SFA and BRUIDS methods, respectively.

The rapid development of multicopters has led to many security problems. In order to prevent multicopters from invading restricted areas or famous buildings, an Anti-UAV Defense System (AUDS) has been developed and become a research topic of interest. Topics under research in relation to this include electromagnetic interference guns for unmanned aerial vehicles (UAVs), high-energy laser guns, US military net warheads, and AUDSs with net guns. However, these AUDSs use either manual aiming or expensive radar to track UAVs. This paper proposes a dual-axis rotary platform with UAV automatic tracking. The tracking platform uses visual image processing technology to track and lock the dynamic displacement of a UAV. When a target UAV is locked, the system uses a nine-axis attitude meter and laser rangefinders to measure its flight altitude and calculates its longitude and latitude coordinates through sphere coordinates to provide UAV monitoring for further defense or attack missions. Tracking tests of UAV flights in the air were carried out using a DJI MAVIC UAV at a height of 30 meters to 100 meters. It was set up for UAV image capture and visual recognition for tracking under various weather conditions by a thermal imager and a full-color camera respectively. When there was no cloud during the daytime, the images captured by the thermal imager and full-color camera provide a high-quality image recognition result. However, under dark weather, black clouds will emit radiant energy and seriously affect the capture of images by a thermal imager. When there is no cloud at night, the thermal imager performs well in UAV image capture. When the UAV is tracked and locked, the system can effectively obtain the flight altitude and longitude and latitude coordinate values.

Precision agriculture is a farm management technology that involves sensing and then responding to the observed variability in the field. Remote sensing is one of the tools of precision agriculture. The emergence of small unmanned aerial vehicles (sUAV) have paved the way to accessible remote sensing tools for farmers. This paper describes the comparison of two popular off-the-shelf sUAVs: 3DR Iris and DJI Phantom 2. Both units are equipped with a camera gimbal attached with a GoPro camera. The comparison of the two sUAV involves a hovering test and a rectilinear motion test. In the hovering test, the sUAV was allowed to hover over a known object and images were taken every second for two minutes. The position of the object in the images was measured and this was used to assess the stability of the sUAV while hovering. In the rectilinear test, the sUAV was allowed to follow a straight path and images of a lined track were acquired. The lines on the images were then measured on how accurate the sUAV followed the path. Results showed that both sUAV performed well in both the hovering test and the rectilinear motion test. This demonstrates that both sUAVs can be used for agricultural monitoring.

Cetacean Morbillivirus, the most relevant pathogen impacting the health and conservation of cetaceans worldwide, has shown in recent years an increased tendency to cross “interspecies barriers”, thereby giving rise to disease and mortality outbreaks in free-ranging dolphins and whales. The present article deals with the evolutionary “trajectories” of this viral pathogen, likely originating from Rinderpest Virus, along with its “journey” from land to sea (and viceversa), mimicking that of cetaceans' terrestrial ancestors.

The “stem cells” discipline represents one of the most dynamic areas in biomedicine. While adult marine/aquatic invertebrate stem cell (MISC) biology is of prime research and medical interest, studies on stem cells from organisms outside the classical vertebrate (e.g., human, mouse, zebrafish) and invertebrate (e.g., Drosophila, Caenorhabditis) models have not been pursued vigorously. Marine/aquatic invertebrates constitute the largest biodiversity and the widest phylogenetic radiation on Earth, from morphologically simple organisms (e.g. sponges, cnidarians), to the more complex mollusks, crustaceans, echinoderms and protochordates. These organisms illustrate a kaleidoscope of MISC-types that participate in the production of a large number of novel bioactive-molecules, many of which are of significant potential interest for human health. MISCs further participate in aging and regeneration phenomena, including whole-body regeneration. For years, the European MISC-community has been highly fragmented and scarce ties were established with biomedical industries in attempts to harness MISCs for human welfare. Thus, it is important to: i) consolidate the fragmented European community working on MISCs; ii) promote and coordinate European research on MISC biology; iii) stimulate young researchers to embark on research in MISC-biology; iv) develop, validate, and network novel MISC tools and methodologies; v) establish the MISC discipline as a forefront interest of biomedical disciplines, including nanobiomedicine; vi) establish collaborations with industries to exploit MISCs as sources of bioactive molecules. In order to fill the recognised gaps, the EC-COST Action 16203 “MARISTEM”, has recently been launched. At its initial stage the consortium unites 26 scientists from EC countries, Cooperating countries and Near Neighbor Countries.

After a local pollution control plant significantly reduced phosphorus loading into a phytoplankton-dominated tributary of the Potomac River in the early 1980’s, water quality and biological communities were monitored bi-weekly from April-September. After a 10-year time-lag, submerged aquatic vegetation (SAV), once abundant in this freshwater tidal embayment, returned to the area in 1993. After additional reductions in nitrogen load starting in 2000, the system completely switched to a SAV-dominated state in 2005. Fish abundance didn’t change during these distinct phase changes, but the fish community structure did. Increases in SAV provided refuge and additional spawning substrate for species with adhesive eggs such as Banded Killifish (Fundulus diaphanus), which is now the most abundant species in the embayment. Other changes seen were a decrease in the relative contribution of open water dwelling species such as White Perch (Morone americana), and an increase of visual predators such as Largemouth Bass (Micropterus salmoides). The 30-year record of data from this Potomac River tributary has revealed many important long-term trends that validate the effectiveness of initiatives to improve water treatment, and will aid in the continued management of the watershed and point source inputs.

Mercury contamination in the Amazon is a significant public health concern arising from both natural sources and intensive mining activities in the region. This metal is used to separate Au from sediments, forming an amalgam that facilitates the extraction of precious metals. Accordingly, this study aimed to address the following question: What is the impact of mining on mercury contamination in the animal and human populations of the Amazon? The overall objective of this study was to conduct a systematic review of the existing literature to identify gaps in geographic coverage and assess the impact of Hg contamination on human and animal populations in the Amazon region. Our hypothesis was that higher mercury concentrations in tissue samples would be found closer to the mining areas. We employed the PECO and PRISMA-ScR protocols to collect articles published between 2017 and 2023, selected based on criteria representing projected points on a map within the biogeographic boundaries of the Amazon. Mercury concentrations increase with trophic levels, reaching high values of 3.7 µg/g in the muscles of predatory fish and 34.9 µg/g in human hair. The average level of mercury in humans exceeds 6 µg/g, surpassing tolerance levels. Although mining regions show high concentrations of Hg, the highest incidence was observed among populations with fish-based diets. Research on fish is crucial for assessing the risk to human health, given that fish are the main source of protein in the region.

Abstract: Little is known about landbird group dynamics in response to predation. Here, we describe the interactions between the Helmeted Guineafowl (Numida meleagris) and its predators regarding the survival advantages grouping behaviour may provide. Livestream webcam observations were conducted in Madikwe Game Reserve (South Africa) from August 2020 to December 2021. Emphasis was placed on predator–prey interaction and its effect on group size and structure in a spatial framework. Although Black-Backed Jackals (Lupulella mesomelas) are not adept at hunting birds, they might negatively impact guineafowl by curtailing their feeding time during winter with constant harassment. We found that these canids often approach larger and denser groups in the morning and late evenings. Surprisingly, guineafowl most often responded by standing nearby. On the contrary, a higher predator-induced response was associated with the presence of predatory birds, arguably as part of an evolutionary stable prey-predator system. This line of thought is also supported by the lack of a precise defensive mechanism elicited by jackal attacks, unlike those of aerial predators. Nevertheless, we contend that grouping in Helmeted Guineafowl is a defensive strategy to respond to predator attacks while individuals engage in daily activities.

Due to the rapid development of unmanned aerial vehicles (UAVs) communication technology, UAVs are gradually competing with primary users (PUs) for spectrum resources. Cognitive radio (CR) is a promising solution to meet the needs. Cooperative spectrum sensing (CSS) is considered as an effective method to detect the PU signal and identify available spectrum resources for UAVs in a cognitive UAV network (CUAVN). However, the cooperative mode among multiple UAVs may incur a high overhead, resulting in performance degradation. Therefore, we introduce a differential sequential 1 (DS1), which incorporates a differential mechanism and leverages the sequential idea based on the classical voting rule to enhance cooperative performance and efficiency. In view of this, we formulate three scenarios to characterize the PU activity and introduce a multi-slot cooperative mode within a single UAV to realize cooperative gain. Further, only the information change about the PU status is sequentially calculated in DS1, and combined with a sequential idea, the efficiency of the voting rule is greatly improved. Moreover, the application of support vector machines (SVM) in dynamic selection enables the selection of the most suitable voting rule based on diverse sensing parameters. This dynamic selection process ensures optimal performance and efficiency by adapting the voting rule to the specific characteristics of the given scenario. Finally, simulation results demonstrate that the superiority of our proposal with respect to the detection performance, sample size and the energy efficiency is evident, which proves the high performance of the proposed policy.

Accurate detection of forest restoration sites using deep learning is crucial for effective ecosystem management, optimizing reforestation efforts, and ensuring environmental sustainability. While various deep learning-based systems have been developed for this purpose, the influence of different neural network architectures on model performance during the identification of forest restoration sites remains underexplored. This study aimed to address this gap by exploring the optimal methodology for extracting and classifying candidate sites for forest restoration based on Convolutional Neural Networks (CNNs), specialized for image recognition. Four categories—arable land, road and barren, quarry, and forest—were defined as candidate sites for forest restoration. A dataset comprising 17,043 samples was divided into training (11,929) and validation (5,114) sets at a ratio of 7:3 for model learning. Model accuracy was evaluated using pixel accuracy (PA) and mean intersection over union (mean IoU). The Resnet152-U-Net model demonstrated a pixel accuracy of 95.2% and a mean IoU of 61.3%, indicating excellent performance in extracting candidate sites for forest restoration. This approach offers spatial and temporal advantages over conventional field surveys or aerial image-based assessments, potentially serving as valuable data for selecting future forest restoration sites.

The vehicular ad hoc network (VANET) when augmented with internet is called internet of vehicles (IoV). In the IoV, vehicles have on board units (OBUs), use communication technologies and customized software to communicate with infrastructure around them. The vehicles can have permanent connectivity and global network awareness based on cloud computing. IoV technologies are still being assessed to ensure that the network is protected from attacks, supports a dynamic vehicular mobility scenarios and meets the low latency communication requirements between the vehicles and the surrounding infrastructure. To meet these strict Quality of Service (QoS) demands, Unmanned Aerial Vehicles (UAVs) play a vital role in IoV ecosystems. An UAV can serve as a mobile roadside unit (mRSU) to transmit data between vehicles, stationery RSUs and other UAVs. The UAVs have better line of sight (LOS) connections which result in better channel states and lower path loss than fixed RSUs. Although the inclusion of UAVs in vehicular networks has improved the system performance by providing better resource management and routing solutions, the traditional optimization techniques with their limitations are often not applicable. Therefore, in recent years, research community is integrating Artificial Intelligence (AI) and Machine Learning (ML) into UAV based IoV ecosystem to better manage resource allocation, routing and mobility management issues to optimize the overall network performance. In this survey paper we review the existing research done in AI/ML based UAV-IoV networks and communications with focus on resource management and routing from year 2019 to 2023. We have studied different AI techniques, their training attributes and architectures in various research works. The limitations of AI-based approaches in terms of required computational resources, availability of real world data and AI models’ complexity for the UAV-IoV environment are also discussed. Finally, the future directions for UAV-IoV research opportunities that can leverage the full potential of AI are presented.

This research focuses on the development of an automated medication delivery system using Unmanned Aerial Vehicle (UAV) Technology and the Internet of Things (IoT) to support telemedicine and community welfare in the KLU region. The system is designed to improve the quality of medical services and medication management for the community in the area. One of the main challenges faced is how the KLU community responds and accepts automatic medication delivery using UAVs and IoT. To overcome this challenge, this research suggests providing education and training to the community about the benefits and usage of the system, as well as displaying a working simulation of the system to the community.

Autonomous target recognition (ATR) plays a crucial role in maintaining situational awareness during environmental monitoring. Unmanned aerial vehicles (UAVs) equipped with autonomous target recognition technology can gather and analyze real-time information about targets, including their locations, sizes, and types. However, UAV-captured images in complex real-world environments often display significant variations in perspective and scale due to changes in UAV altitude and distance. Existing methods for autonomous target recognition on UAVs struggle to capture targets from large field-of-view and multi-scale images, resulting in low recognition accuracy and high false-positive rates. This paper introduces two novel Slimmable neural network models, namely SE-YOLOv5s and ST-YOLOv5s, which are based on the YOLOv5s architecture. These models incorporate the Squeeze and Excitation and Swin-Transformer mechanisms to enhance the ability to extract features from large field-of-view and multi-scale images. To evaluate their performance, experiments were conducted on the Visdrone19 aerial dataset. Compared to the state-of-the-art YOLOv5s algorithm, the utilization of SE-YOLOv5s and ST-YOLOv5s for autonomous target recognition on low-altitude drones resulted in improvements in both accuracy and false-positive rates. These proposed methods combine Slimmable neural network design with feature enhancement mechanisms, addressing the challenges posed by complex real-world environments in UAV missions. The advancements in autonomous target recognition on low-altitude drones will significantly contribute to enhancing situational awareness in future environmental monitoring.

Urbanization and climate change have increased the need for stormwater management and nature-based solutions. Decisions made at project level impact the emergence of systemic traits of the stormwater network and the functionality of the catchment areas in urban planning. To that end, it is vital to introduce the decision-making tools for analyzing both the utilities and amenities of nature-based solutions (NBS) to increase their adoption to reduce the peak loads in the stormwater system, and to that end, mitigate the impacts of climate change. This paper demonstrates a workflow using drone-based photogrammetry, 3D modeling, and simulation software to generate visual and functional models assisting in informed decision-making in the design of stormwater systems as functional landscape architecture. Using aerial data from drones and modeled design solutions, the proposed workflow simulates rain events, infiltration, evaporation, water flow, and accumulation of stormwater in a way that allows the visual and quantified analysis of detailed landscape architecture designs. The paper provides an example of a rooftop site simulation demonstrating the infiltration and flow of water to the drainage. The visual decision-making method provided can aid in investment decisions for functional landscape design in support of stormwater management.

This paper presents the design of a digital twin that blends aviation, gaming, simulation, and Geographic Information Systems (GIS) to create a synthetic environment within which strategies, laws and platforms for electric aviation may be tested out. This digital twin has been called Future Urban Synthetic Environment (FUSE). FUSE includes an in-built Unmanned Traffic Management (UTM) that can be used to run simulations to test the coordination of all urban traffic. It uses real GIS tagged imagery data and implements it at runtime into the game engine and thus link the optimised imagery loading to the visual performance of the simulation engine. FUSE provides a 3D digital twin of specified areas designed to simulate the effects (in terms of noise, visual impact, privacy) of drones and electric air taxis operating under various operational scenarios (such as number of deliveries allowed per day, maximum payload weight, no-fly areas, position of depots, vertiports, etc.). With so much high-fidelity data it is difficult for any game system to effectively render the environment and do justice to the detail whilst enabling enough of the landscape to be rendered to keep in focus the detail when looking out at the horizon.

Access to broadband communications in different parts of the world has become a priority for some governments and regulatory authorities around the world in recent years. Building new digital roads and pursuing a connected society includes looking for easier access to the Internet. In general, not all the areas where people congregate are fully covered, especially in rural zones, thus restricting access to data communications and bringing inequality. In the present review article, we have comprehensively surveyed the use of three platforms to deliver broadband services to such remote and low-income areas are proposed: Unmanned Aerial Vehicles (UAV), Altitude Platforms (APS), and Low-Earth Orbit (LEO) satellites. These novel strategies support the connected and accessible world hypothesis. Hence UAVs are considered a noteworthy solution since their efficient maneuverability can aboard the rural coverage issues or not-spots.

Αbstract: To date, countless satellite image fusions have been made, mainly with panchromatic spatial resolution to a multispectral image ratio of 1/4, fewer fusions with lower ratios, and relatively recently fusions with much higher spatial resolution ratios have been published. Apart from this, there is a small number of publications studying the fusion of aerial photographs with satellite images, with the year of image acquisition varying and the dates of acquisition not mentioned. In addition, in these publications, either no quantitative controls are performed on the composite images produced, or the aerial photographs are recent and colorful and only the RGB bands of the satellite images are used for data fusion purposes. The objective of this paper is the study of the addition of multispectral information from satellite images to black and white aerial photographs of the 80s decade (1980–1990) with small difference (just a few days) in their image acquisition date, the same year and season. Quantitative tests are performed in two case studies and the results are encouraging, as the accuracy of the classification of the features and objects of the Earth’s surface is improved and the automatic digital extraction of their form and shape from the archived aerial photographs is now allowed. This opens up a new field of use for the black and white aerial photographs and archived multispectral satellite images of the same period in a variety of applications, such as the temporal changes of cities, forests and archaeological sites.

Riparian restoration is an important objective for landscape managers seeking to redress the widespread degradation of riparian areas and the ecosystem services they provide. This study investigated the long-term outcomes of ‘one-off’ restoration activities undertaken in the Upper Murrumbidgee Catchment, NSW, Australia. The objective of the restoration was to protect and enhance riparian vegetation and control erosion, and consequently reduce sediment and nutrient delivery into the Murrumbidgee River. To evaluate the outcomes 10 years after restoration, rapid riparian vegetation and geomorphological assessments were undertaken at 29 sites spanning the four different restoration methods used (at least five replicates per treatment), as well as at nine comparable untreated sites. We also trialed the use of aerial imagery to compare width of riparian canopy vegetation and projective foliage cover prior to restoration with that observed after 10 years. Aerial imagery demonstrated the width of riparian canopy vegetation and projective foliage cover increased in all restored sites, especially those with native plantings. The rapid assessment process indicated that 10 years after riparian restoration, the riparian vegetation was in a better condition at treated sites compared to untreated sites. Width of riparian canopy vegetation, native mid-storey cover, native canopy cover and seedling recruitment were significantly greater in treated sites compared to untreated sites. Geomorphological condition of treated sites was significantly better than untreated sites, demonstrating the importance of livestock exclusion to improve bank and channel condition. Our findings illustrate the value of ‘one-off’ restoration activities in achieving long-term benefits for riparian health. We have demonstrated that rapid assessments of the vegetation and geomorphological condition can be undertaken post-hoc to determine the long-term outcomes, especially when supported with analysis of historical aerial imagery.

Wetlands benefits can be summarized but are not limited to their ability to store floodwaters and improve water quality, providing habitats for wildlife and supporting biodiversity, as well as aesthetic values. Over the past few decades, remote sensing and geographical information technologies has proven to be a useful and frequent applications in monitoring and mapping wetlands. Combining both optical and microwave satellite data can give significant information about the biophysical characteristics of wetlands and wetlands` vegetation. Also, fusing data from different sensors, such as radar and optical remote sensing data, can increase the wetland classification accuracy. In this paper we investigate the ability of fusion two fine spatial resolution satellite data, Sentinel-2 and the Synthetic Aperture Radar Satellite, Sentinel-1, for mapping wetlands. As a study area in this paper, Balikdami wetland located in the Anatolian part of Turkey has been selected. Both Sentinel-1 and Sentinel-2 images require pre-processing before their use. After the pre-processing, several vegetation indices calculated from the Sentinel-2 bands were included in the data set. Furthermore, an object-based classification was performed. For the accuracy assessment of the obtained results, number of random points were added over the study area. In addition, the results were compared with data from Unmanned Aerial Vehicle collected on the same data of the overpass of the Sentinel-2, and three days before the overpass of Sentinel-1 satellite. The accuracy assessment showed that the results significant and satisfying in the wetland classification using both multispectral and microwave data. The statistical results of the fusion of the optical and radar data showed high wetland mapping accuracy, with an overall classification accuracy of approximately 90% in the object-based classification. Compared with the high resolution UAV data, the classification results give promising results for mapping and monitoring not just wetlands, but also the sub-classes of the study area. For future research, multi-temporal image use and terrain data collection are recommended.

Increased salinity caused by saltwater intrusion or runoff from de-icing salts can severely affect freshwater vegetation and deteriorate aquatic ecosystems. These habitats can be restored with freshwater ecotypes (locally adapted populations) that tolerate above-normal salinity. Vallisneria americana is a prominent species in many freshwater ecosystems that responds differently to abiotic conditions such as substrate composition and fertility, so in this study we evaluated the effects of salt stress on 24 ecotypes of V. americana. Instant Ocean aquarium salt was used to create saline solutions [0.2 to 20.0 parts per thousand (ppt)], then plants were abruptly exposed to these solutions and maintained in these concentrations for 5 weeks before being visually assessed for quality and destructively harvested. Analysis of variance and non-linear regression were used to calculate LC₅₀ values – the lethal concentration of salt that reduced plant biomass and quality by 50% compared to control treatment. Growth rate and visual quality varied significantly among ecotypes, and ecotypes that were most and least sensitive to salt had 50% biomass reductions at 0.47 and 9.10 ppt, respectively. All ecotypes survived 10.0 ppt salinity concentration but none survived at 20.0 ppt, which suggests the maximum salinity concentration tolerated by these ecotypes is between 15.0 and 20.0 ppt.

This study introduces a fruit harvesting mechanism powered by a single motor, designed for integration with unmanned aerial vehicles (UAVs). The mechanism performs reciprocating motion by converting linear motion into rotational motion. Consequently, the end-effector can execute multi-dimensional kinematic trajectories, including biaxial and rotational movements, synchronized with the motor's position. These axial and rotational motions facilitate the gripper's ability to reach, retrieve, and detach fruit from branches during the harvesting process. Notably, a critical consideration in designing this fruit harvesting mechanism is to generate the necessary torque at the end-effector while minimizing reaction forces and torque that could destabilize the UAV during flight. With these considerations in mind, this preliminary study aimed to harvest a Fuji apple and conducted dynamic analysis. We constructed a prototype of the single motor-driven fruit harvesting mechanism using a suitable servo motor. To assess its mechanical performance and evaluate its impact on the hexacopter, we developed both a specific test platform featuring a six-spherical-prismatic-spherical parallel structure and a virtual environmental flight simulator. Overall, the results demonstrate the successful harvesting of a Fuji apple weighing approximately 300 g by the single motor-driven fruit harvesting mechanism, with no adverse effects observed on the hexacopter's operation.

Abstract: ‘Temperate movements: Artistic Responses to Climate Vulnerability in the Canadian Arctic’ uses examples and methodologies from contemporary art and theory to think through the impact and legacy of Arctic Sea ice melt upon Inuit culture. Here, the work of Inuit artists Eldred Allen and Maureen Gruben illustrates how melting Arctic Sea ice maybe considered as relational space best understood through a cultural, rather than an exclusively scientific and data-oriented relationship to global sea rise level. Such a cultural perspective, which emphasizes the interconnection between humans, animals, and the environment, can counter dominant climate change imagery which tends to homogenize and de-socialize the Arctic as a single concept. The work of these Inuit artists is also considered as outreach to non-Inuit communities, both in terms of photographic art forms circulating in ‘southern’ contemporary art contexts, as well as becoming ‘novel’ indexes of climate vulnerability and resilience.

This paper presents a study on the influence of the frequency variation of a nonlinear1 control allocation technique execution, developed by the author [1], named by Fast Control2 Allocation (FCA) for the Quadrotor Tilt-Rotor (QTR) aircraft. Then, through Software In The3 Loop (SITL) simulation, the proposed work considers the use of Gazebo, QGroundControl, and4 Matlab applications, where different frequencies of the FCA can be implemented separated in5 Matlab, always analyzing the QTR stability conditions from the virtual environment performed in6 Gazebo. TheresultsshowedthattheFCAneedsatleast200HzoffrequencyfortheQTRsafeflight7 conditions, i. e., 2 times smaller than the main control loop frequency, 400 Hz. Lower frequencies8 than this one would case instability or crashes during the QTR Operation.

The current study investigated the use of two-dimensional spatial distribution mapping representing the chlorophyll-a level in a river generated via an unmanned aerial vehicle (UAV) and an unmanned surface vehicle (USV). A domestically developed UAV (Remo-M, Uconsystem Inc., Korea) and a USV developed by our research team were used to collect data from the Nae Seong stream in Korea. An adaptation of the “Data Cleaner” tool was developed and used for USV data processing and analysis. The operation of the autonomous USV was successful. Four previously described indices for quantifying algal blooms in rivers were utilized to create chlorophyll-a images, the normalized difference vegetation index (NDVI), the normalized green red difference index, the green normalized difference vegetation index (GNDVI), and the normalized difference red edge index. The suitability of the linear regression analysis of the correlation between the spectral indices obtained using the UAV and the in situ chlorophyll-a data obtained using the USV was evaluated with the coefficient of determination (R2) at a significance level of p &lt; 0.001. In field application and correlational analysis the NDVI was strongly correlated with chlorophyll-a (R2 = 0.88, p < 0.001), and the GNDVI was moderately correlated with chlorophyll-a (R2 = 0.74, p < 0.001). The map of chlorophyll-a was successfully quantified using the UAV and USV hybrid platforms.

In recent years, weeds is responsible for most of the agricultural yield losses. To deal with this threat, farmers resort to spraying pesticides throughout the field. Such method not only requires huge quantities of herbicides but impact environment and humans health. One way to reduce the cost and environmental impact is to allocate the right doses of herbicide at the right place and at the right time (Precision Agriculture). Nowadays, Unmanned Aerial Vehicle (UAV) is becoming an interesting acquisition system for weeds localization and management due to its ability to obtain the images of the entire agricultural field with a very high spatial resolution and at low cost. Despite the important advances in UAV acquisition systems, automatic weeds detection remains a challenging problem because of its strong similarity with the crops. Recently Deep Learning approach has shown impressive results in different complex classification problem. However, this approach needs a certain amount of training data but, creating large agricultural datasets with pixel-level annotations by expert is an extremely time consuming task. In this paper, we propose a novel fully automatic learning method using Convolutional Neuronal Networks (CNNs) with unsupervised training dataset collection for weeds detection from UAV images. The proposed method consists in three main phases. First we automatically detect the crop lines and using them to identify the interline weeds. In the second phase, interline weeds are used to constitute the training dataset. Finally, we performed CNNs on this dataset to build a model able to detect the crop and weeds in the images. The results obtained are comparable to the traditional supervised training data labeling. The accuracy gaps are 1.5% in the spinach field and 6% in the bean field.

TVs and monitors are among the most widely used displays in various environments. However, they have limitations in their physical display conditions, such as a fixed size/position and a rigid/flat space. In this paper, we suggest a new "Display In the Wild" (DIW) concept to overcome the aforementioned problems. Our proposed DIW system allows us to display a flexibly large screen on dynamic non-planar surfaces at an arbitrary display position. To implement our DIW concept practically, we choose a projector as the hardware configuration in order to a generate screen anywhere with different sizes. However, distortion occurs when the projector displays content on a surface that is dynamic and/or non-planar. Therefore, we propose a distortion correction method for DIW to overcome the aforementioned surface-constraints. Since projectors are not capture devices, we propose using a depth camera to determine the distortions on the surfaces quickly. We also propose DIW-specific calibration and fast/precise correction methods. Our calibration method is constructed to easily and quickly detect the projection surface, and also allows our proposed system to accommodate the intrinsic parameters such as a display resolution and field of view. We accomplish a fast undistortion process of the projector by considering only surface boundary pixels, which enables our method to run in real time. In our comprehensive experiments, the proposed DIW system generates undistorted screens such as TVs and monitors on dynamic non-planar surfaces at an arbitrary display position with Unmanned Aerial Vehicles (UAVs) in a fast and accurate manner, demonstrating its usefullness in practical DIW scenarios.

Autonomous aerial refueling technology is an effective solution to extend flight duration of unmanned aerial vehicles, and also a great challenge due to its high risk. For autonomous probe-and-drogue refueling tasks, relative navigation to provide relative position between the receiver aircraft and the refueling drogue is the first and essential step, and vision-based method is the most frequently used. A new monocular vision navigation sensor with image processing strategy consisting of the drogue detection method and the tracking method is developed for autonomous aerial refueling in this paper. In the drogue detection method, thresholding and mathematical morphology method are adopted to eliminate image interference, and contours extraction method is applied to obtain all contours, which are then subsequently checked to achieve target contour of drogue. In the tracking method, a rectangle of interest (ROI) of current frame image is determined by positioning results of last frame, and then processed by the previous drogue detection method. Finally, the proposed image processing strategy in monocular vision navigation sensor is validated using real flight images, which are captured from an autonomous aerial refueling testbed using a micro six-rotor aircraft as receiver aircraft.

Remotely Piloted Aircraft Systems (RPAS) are increasingly becoming relevant actors flying through the airspace and will assume much more importance in the future perspective. In order to allow their safe integration with manned conventional traffic in non-segregated airspaces, in accordance with the overall Air Traffic Management (ATM) paradigm, specific enabling technologies are needed. As well known, among the enabling technologies identified as crucial for RPAS integration into the overall ATM system, the Detect and Avoid (DAA) technology is fundamental. In the meantime, to support extended surveillance, the universal introduction on-board of aircraft of cooperative Automatic Dependent Surveillance – Broadcast (ADS-B) is increasingly implemented, having the potential to allow coverage of the whole airspace also in remote areas not usually covered by conventional radar surveillance. In this paper, the experimental results are presented and discussed that have been obtained through the real-time validation, with hardware and human in the loop (RTS-HIL) simulations, of an automatic ADS-B based Separation Assurance and Collision Avoidance System aimed to support RPAS automatic operations as well as remote pilot decision making. In the paper, after an introductory outline of the Concept of Operations (ConOps) of the system and of its architectural organization, while also providing basic information about the main system functionalities, the description is reported of the tests that have been carried out and the obtained results are described and discussed, in order to emphasize the performances and limitations of the proposed system. In particular, not only the quantitative performances obtained are reported and commented but also the feedbacks received by the pilots in order to improve the system are described, for instance in terms of preferred typology of conflict resolution manoeuver elaborated by the system.

Eight amniote genera (representing four clades) became aquatic during the Permian. The four clades were mesosaurids, tangasaurids, the neodiapsid Claudiosaurus, and the procolophonid Barasaurus. Two of eight genera survived the end-Permian mass extinction, but did not last long into the Mesozoic. A previously undescribed specimen of Claudiosaurus germaini, preserved in a lacustrine concretion from the Sakamena Formation, Madagascar, bears seventeen vertebrae that has been split along an approximate horizontal plane to reveal sections of neural canal casted in white calcite. Enlargement of the neural canal in the sacral region of this specimen of Claudiosaurus (vertebral segments 22-26) is more similar to that of Tupinambis (segments 24-28) than it is to Testudo (segments 16-23). Claudiosaurus skeletal anatomy provides evidence for swim propulsion by both hind limbs and by undulation of a dorsal-ventrally flattened tail. Evidence for the latter includes elongate transverse processes on distal tail vertebrae. Other Permian aquatic reptile genera (Mesosaurus, Hovasaurus, Barasaurus) used snake-like side-to-side tail undulation, whereas Claudiosaurus used cetacean-like up-down tail undulation in the vertical plane. It seems unlikely that any of these animals were particularly fast swimmers.

Unmanned aerial vehicles (UAVs) have increasingly become integral to multi-access edge computing (MEC) due to their flexibility and cost-effectiveness, especially in the B5G and 6G eras. This paper aims to enhance the Quality of Experience (QoE) in large-scale UAV-MEC networks by minimizing the shrinkage ratio through optimal decision-making in computation mode selection for each user device (UD), UAV flight trajectory, bandwidth allocation, and computing resource allocation at edge servers. However, the interdependencies among UAV trajectory, binary task offloading mode, and computing/network resource allocation across numerous IoT nodes pose significant challenges. To address these challenges, we formulate the shrinkage ratio minimization problem as a mixed-integer nonlinear programming (MINLP) problem and propose a two-tier optimization strategy. To reduce the scale of the optimization problem, we first design a low-complexity UAV partition coverage algorithm based on the Welzl method and determine the UAV flight trajectory by solving a Traveling Salesman Problem (TSP). Subsequently, we develop a coordinate descent (CD) based method and an alternating direction method of multipliers (ADMM) based method for network bandwidth and computing resource allocation in the MEC system. The CD-based method is simple to implement and has a low computational complexity, while the ADMM-based method can further enhance the optimization result through joint optimization. Extensive simulations demonstrate that our proposed algorithms perform well in large-scale edge networks and outperform other representative benchmark 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, \eg{} 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, Luoyang Pedestrian Dataset \footnote{The dataset can be downloaded through this anonymous link: \url{https://drive.google.com/drive/folders/13po1FX7Qk5RDgb60-dzOi74cq_kqDJtM?usp=sharing} }, 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 Luoyang Pedestrian Dataset, our proposed MSA-YOLO algorithm significantly outperform the most commonly 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.)

The rise of the Internet of Things (IoT) has opened up exciting possibilities for new applications. One such novel application is the modernization of maritime communications. Effective maritime communication is vital for ensuring the safety of crew members, vessels, and cargo. The maritime industry is responsible for the transportation of a significant portion of global trade, and as such, the efficient and secure transfer of information is essential to maintain the flow of goods and services. With the increasing complexity of maritime operations, technological advancements such as unmanned aerial vehicles (UAVs), autonomous underwater vehicles (AUVs), and the Internet of Ships (IoS) have been introduced to enhance communication and operational efficiency. However, these technologies also bring new challenges in terms of security and network management. Compromised IT systems, with escalated privileges, can potentially enable easy and ready access to operational technology (OT) systems and networks with the same privileges, with an increased risk of zero-day attacks. In this paper, we first provide a review of the current state and modalities of maritime communications. We then review the current adoption of Software-Defined Radios (SDRs) and Software-Defined Networks (SDNs) in the maritime industry and evaluate their impact, as IoT enablers, on the future of maritime industry operations, safety, and security.

In recent years, unmanned aerial vehicles (UAVs) have become a valuable platform for many applications, including communication networks. UAV-enabled wireless communication faces challenges in complex urban and dynamic environments. UAVs can suffer from power limitations and path losses caused by non-line-of-sight connections, which may hamper better communication performance. To address these issues, reconfigurable intelligent surfaces (RIS) have been proposed as a helpful tool to enhance UAV communication networks. However, due to the high mobility of UAV, complex channel environments, and dynamic RIS configurations, it is challenging to estimate the link quality of ground users. In this paper, we propose a link quality estimation model using a gated recurrent unit (GRU) to assess the link quality of ground users for a multi-user RIS-assisted UAV-enabled wireless communication system. Our proposed framework uses a time series of user channel data and RIS phase shift information to estimate the quality of the link for each ground user. Simulation results showed that the proposed GRU model effectively and accurately estimates the link quality of the ground users in the RIS-assisted UAV-enabled wireless communication network.

Temporary streams are a key component of the hydrological cycle in arid and semi-arid regions, but their flow is highly variable and difficult to measure. In this paper, we present a novel approach that could be used to assess the flow of temporary streams and characterize their environmental status. Specifically, we apply the Image Velocimetry method to estimate surface velocity in temporary streams using unmanned aerial vehicles (UAVs) equipped with optical sensors (IV-UAV method). The IV-UAV method enables the easy, safe and quick estimation of the velocity on the water’s surface. This method was applied in different temporary streams in Lesvos Island, Greece. The produced results show that the IV-UAV can be implemented at low discharges, at temporary streams and at small streams. However, there were many occasions that measurements were unable due to various reasons such as dense vegetation or archaeological sites. Despite of this, the proposed methodology could be incorporated in optical protocols which are used to assess the environmental status of temporary streams of Mediterranean conditions. Finally, this would become a valuable tool for understanding the dynamics of these eco-systems and monitoring changes over time.

This paper proposes a high-performance and low-cost in-flight collision avoidance system based on background subtraction for unmanned aerial vehicles (UAVs). The pipeline of proposed in-flight collision avoidance system is as follows: (i) dynamic background subtraction to remove the background and to detect moving objects, (ii) denoise using morphology and binarization methods, (iii) Euclidean clustering to cluster the moving object and to remove noise blobs, (iv) distinguish independent objects and track the movement using Kalman filter, and (v) collision avoidance using proposed decision-making techniques. This work focuses on the design and the demonstration of a vision-based fast moving object detection and tracking system with decision-making capabilities to perform evasive maneuvers to replace high vision system such as event camera. The development of high-performance, low-cost unmanned aerial vehicles paired with rapid progress in vision-based perception systems herald a new era of autonomous flight systems with mission-ready capabilities. One of the key features of an autonomous UAV is a robust mid-air collision avoidance strategy. The novelty of our method lies in the motion-compensating moving object detection framework, which accomplishes the task with background subtraction via 2-D transformation approximation. Clustering and tracking algorithms process detection data to track independent objects, and stereo-camera-based distance estimation is conducted to estimate the 3-D trajectory, which is then used during decision-making procedures. The examination of the system is conducted with a quadrotor UAV test vehicle, and appropriate algorithm parameters for various requirements are deduced.

This paper proposes a compact planar Wi-Fi antenna with optimized radiation patterns for small uncrewed aerial vehicle (UAV) applications in both urban and open areas. It is suitable for mounting on the outermost side of the non-metallic wing of small UAVs. It has small dimensions of 16.5 mm (L) by 30.3 mm (W) by 1.6 mm (h), and the measured results of its prototype are in agreement with simulated data. Its impedance bandwidths over the two frequency ranges are 2.11 to 2.58 GHz and 5.06 to 7.5 GHz ( ). At 5.8 GHz, it has stronger radiation below the small UAV to reduce interference from rare-use directions. Its maximum radiations, the directions of the maximum radiation in each elevation plane, are below the UAV and between 14° and 29° from the horizontal plane. At 2.4 GHz, it has quasi-omnidirectional radiation to ensure a stable link in all directions, and its maximum radiations are near the horizontal plane. The optimized radiation patterns at 5.8 and 2.4 GHz can provide more antenna gain when the small UAV flies farther in urban and open areas, respectively. In addition, it has good vertically polarized radiation for long-distance applications.

Quadrotor drones have rapidly gained interest recently. Numerous studies are underway for the commercial use of autonomous drones, and especially the distribution businesses are taking serious reviews on drone delivery services. However, there are still many concerns about urban drone operations. The risk of failures and accidents makes it difficult to provide drone-based services in the real world with ease. There have been many studies that introduced supplementary methods to handle drone failures and emergencies. However, we discovered the limitation of the existing methods. The majority of approaches were improving PID-based control algorithms which is the dominant drone control method. This type of low-level approach lacks situation awareness and the ability to handle unexpected situations. This study introduces an event-based control methodology that takes a high-level diagnosing approach that can implement situation awareness via time-window. While leaving the low-level controller to involve in operating the drone for most of the time in normal situations, our controller operates at a higher level and detects unexpected behaviors and abnormal situations of the drone. We tested our method with real-time 3D computer simulation environments with Unreal Engine[15] and AirSim[31]. We were able to verify that our approach can provide enhanced double safety and better ensure safe drone operations. We hope our discovery to possibly contribute to the advance of real-world drone services in the near future.

Recent expansion in data sharing has created unprecedented opportunities to explore structure-function linkages in ecosystems across spatial and temporal scales. However, characteristics of the same data product, such as resolution, can change over time or spatial locations, as protocols are adapted to new technology or conditions, which may impact the data’s potential utility and accuracy for addressing end user scientific questions. The National Ecological Observatory Network (NEON) provides data products for users from 81 sites and over a planned 30-year time frame, including discrete return Light Detection and Range (LiDAR) from an airborne observatory platform. LiDAR is a well-established and increasingly available remote sensing technology for measuring three-dimensional (3D) characteristics of ecosystem and landscape structure, including forest structural diversity. The LiDAR product that NEON provides can vary in point density from 2 – 25+ points/m² depending on instrument and acquisition date. We used NEON LiDAR from five forested sites to (1) identify the minimum point density at which structural diversity metrics can be robustly estimated across forested sites from different ecoclimatic zones in the USA and (2) to test the effects of variable point density on the estimation of a suite of structural diversity metrics and multivariate structural complexity types within and across forested sites. Twelve out of sixteen structural diversity metrics were sensitive to LiDAR point density in at least one of the five NEON forested sites. The minimum point density to reliably estimate the metrics ranged from 2.0 to 7.5 pt/m^2, but our results indicate that point densities above 7-8 pt/m² should provide robust measurements of structural diversity in forests for temporal or spatial comparisons. The delineation of multivariate structural complexity types from a suite of 16 structural diversity metrics was robust within sites and across forest types for a LiDAR point density of 4 pt/m² and above. This study shows that different metrics of structural diversity can vary in their sensitivity to the resolution of LiDAR data and users of these open-source data products should consider the point density of their data and use caution in metric selection when making spatial or temporal comparisons from these datasets.

A novel adaptive neural network-based fault-tolerant control scheme is proposed for six-degree freedom nonlinear helicopter dynamic. The proposed approach can detect and mitigate sensors' faults in real-time. An adaptive observer-based on neural network (NN) and extended Kalman filter (EKF) is designed, which incorporates the helicopter's dynamic model to detect faults in the navigation sensors. Based on the detected faults, an active fault-tolerant controller, including three loops of dynamic inversion, is designed to compensate for the occurred faults in real-time. The simulation results showed that the proposed approach is able to detect and mitigate different types of faults on the helicopter navigation sensors, and the helicopter tracks the desired trajectory without any interruption.

Aero submarines (aerosubs) are vehicles that can both fly both in air and travel under water. The concept of dual aerial and aquatic vehicles emerged in 1939 when Russian engineer Boris Ushakov proposed the “flying submarine”[1], and this was followed by further developments including RFS1 [2], convair project in 1964[3] , etc. however, to date, limited attempt has been diverted towards the advanced development of such aircraft. This is heavily influenced by challenges associated with the design and operation of the same. Based on the review of literature the authors aim to introduce a theoretical design for an aerosub (QFS-20) with a view to address the design and operation issues including power, entry to and exit from water.

UAV photogrammetry development during the last decade has allowed to catch information at a very high spatial and temporal resolution from terrains with very difficult or impossible human access. This paper deals with the application of these techniques to study and produce information of very extreme topography which is useful to plan works on this terrain or monitoring it over the time to study its evolution. The methodology stars with the execution of UAV flights on the cut slope studied, one with the cam vertically oriented and other at 45º respect that orientation. Ground control points (GCPs) and check points (CPs) were measured for georeference and accuracy measurement purposes. Orthophoto was obtained projecting on a fitted plane to a studied surface. Moreover, since a digital surface model (DSM) is not able to represent faithfully that extreme morphology, information to project works or monitoring it has been derived from the point cloud generated during the photogrammetric process. An informatics program was developed to generate contour lines and cross sections derived from the point cloud, which was able to represent all terrain geometric characteristics, like several Z coordinates for a given planimetric (X, Y) point. Results yield a root mean square error (RMSE) in X, Y and Z directions of 0.053 m, 0.070 m and 0.061 m respectively. Furthermore, comparison between contour lines and cross sections generated from point cloud with the developed program on one hand and those generated from DSM on other hand, shown that the former are capable of representing terrain geometric characteristics that the latter cannot. The methodology proposed in this work has been shown as an adequate alternative to generate manageable information, as orthophoto, contour lines and cross sections, useful for the elaboration, for example, of projects for repairing or maintenance works of cut slopes with extreme topography.

The permanent protection of groundwater and surface sources of drinking water in the required raw water quality from contamination is the primary foundation of preserving the life of society and the entire engineering infrastructure operability not only in the Czech Republic but also in other countries of the world. One of the basic ways of contamination prevention enhancement and risk mitigation of putting a water source out of operation is the identification of the given risk as well as soil contamination monitoring. The purpose of this study is to understand specific aspects of contamination threats to drinking water sources and determine possible solutions to eliminate the risk, shown in the example case study in the Czech Republic conditions. Based on the findings, a proposal to enhance the prevention of water source contamination has been elaborated in the form of recommendations regarding the ways and means used to monitor the risk areas and to implement technical measures so as to eliminate or terminate the specified risk. A complex approach based on the integration of interdisciplinary factors and methodology is required to minimize the threats of aquatic ecosystems contamination in the Czech Republic.

Accurate and timely prediction of Napa cabbage (Brissica rapa subsp. Perkinensis) fresh weight is crucial for optimizing harvest timing, crop management, and supply chain logistics, contributing to food security and price stabilization. Traditional manual sampling methods are labor-intensive and imprecise. This study addresses this challenge by developing a comprehensive (artificial intelligence) AI-powered model for predicting Napa cabbage fresh weight using unmanned aerial vehicle (UAV)-based multi-sensor data. High-resolution RGB, multispectral, and thermal infrared (TIR) imagery were collected over a Napa cabbage field throughout the 2020 growing season. Various vegetation indices, crop features (vegetation fraction, crop height model), and water stress indi-cators (CWSI) were extracted from the imagery. Three AI algorithms—deep neural network (DNN), support vector machine (SVM), and random forest (RF)—were trained and evaluated, with the DNN model consistently outperforming the others. The DNN model achieved the highest accuracy (R² = 0.86 for training, 0.82 for testing; root mean square error (RMSE) = 0.432 kg for training, 0.465 kg for testing) during the mid-to-late rosette growth stage (DAP 35-42), highlighting this period as crucial for fresh weight estimation due to stable leaf area and well-developed canopy structure. The model tended to underestimate the weight of Napa cabbages exceeding 5 kg, potentially due to limited samples and saturation effects of vegetation indices. However, the overall error rate was less than 5%, demonstrating the feasibility and effectiveness of this approach. Spatial analysis revealed that the model accurately captured the variability in Napa cabbage growth across different soil types and irrigation conditions, particularly reflecting the positive impact of drip irrigation on the sandy loam plot. Bias analysis indicated the DNN model's tendency to overestimate smaller Napa cabbages (2 kg), suggesting areas for future refinement. This study demonstrates the potential of UAV-based multi-sensor data and AI algorithms for accurate and non-invasive prediction of Napa cabbage fresh weight. The developed DNN model offers a promising tool for optimizing harvest timing, improving crop management practices, and en-hancing supply chain efficiency. Future research should focus on refining the model for specific weight ranges and diverse environmental conditions, as well as extending its application to other crops, to further advance precision agriculture and contribute to sustainable food production.

Popular geo-computer vision works make use of aerial imager with sizes ranging from 64 × 64 to 1024 × 1024 pixels without any overlap, although the learning process of deep learning models can be affected by the reduced semantic context or the lack of information near image boundaries. In this work, the impact of three image sizes (256 × 256, 512 × 512, and 1024 × 1024 pixels) and two image overlap levels (no overlap and 12.5% overlap) on the performance of road classification models was statistically evaluated. For this, two convolutional neural networks used in various tasks of geospatial object extraction were trained (using the same hyperparameters) on a large dataset (containing aerial image data covering 8650 km² of the Spanish territory that was labelled with binary road information) under twelve different scenarios, each scenario featuring a different combination of tile size and overlap. To assess their generalisation capacity, the performance of all resulting models was evaluated on a data from novel areas covering approximately 825 km². The performance metrics obtained were analysed using appropriate descriptive and inferential statistical techniques to evaluate the impact of the performance at distinct levels of the fixed factors (tile size, tile overlap, neural network architecture). Statistical tests were applied to study the main and interaction effects of the fixed factors on the performance. A significance level of 0.05 was applied to all the null hypothesis tests. The results were highly significant for the main effects (p-values lower than 0.001), while the two-way and three-way interaction effects among them had different levels of significance. The results indicate that the training of road classification models on images with a higher tile size (more semantic context) and tile overlap (additional border context and continuity) significantly impacts their performance. The best model was trained on a dataset featuring tiles with a size of 1024 × 1024 pixels and a 12.5% overlap and achieved a loss value of 0.0984, an F1 score of 0.8728, and an ROC-AUC score of 0.9766, together with an error rate of 3.5% on the test set.

In recent years, the utilization of machine learning algorithms and advancements in Unmanned Aerial Vehicle (UAV) technology have marked significant shifts in remote sensing practices. In particular, the integration of machine learning with physical models and their application in UAV-satellite data fusion have emerged as two prominent approaches for estimating vegetation biochemistry. This study evaluates the performance of five machine learning regression algorithms (MLRAs) for mapping crop canopy chlorophyll at the Kellogg Biological Station (KBS) in Michigan, USA, across three scenarios: (1) application to Landsat 7, RapidEye, and PlanetScope satellite images; (2) application to UAV-satellite data fusion; and (3) integration with the PROSAIL radiative transfer model (hybrid methods PROSAIL + MLRAs). The results indicate that the majority of the five MLRAs utilized in UAV-satellite data fusion perform better than the five PROSAIL + MLRAs. The general trend suggests that the integration of satellite data with UAV-derived information, including the normalized difference red-edge index (NDRE), canopy height model, and leaf area index (LAI), significantly enhances the performance of MLRAs. The UAV-RapidEye dataset exhibits the highest coefficient of determination (R2) and the lowest root-mean-square errors (RMSE) when employing kernel ridge regression (KRR) and Gaussian processes regression (GPR) (R2 = 0.89 and 0.89 and RMSE = 8.99 µg/cm2 and 9.65 µg/cm2, respectively). Similar performance is closely observed for the UAV-Landsat and UAV-PlanetScope datasets (R2 = 0.86 and 0.87 for KRR, respectively). For the hybrid models, maximum performance is attained with Landsat data using KRR and GPR (R2 = 0.77 and 0.51 and RMSE = 33.10 µg/cm2 and 42.91 µg/cm2, respectively), followed by R2 = 0.75 and RMSE = 39.78 µg/cm2 for PlanetScope data upon integrating partial least squares regression (PLSR) into the hybrid model. Across all hybrid models, RapidEye data yielded the most stable performance, with R2 ranging from 0.45 to 0.71 and RMSE ranging from 19.16 µg/cm2 to 33.07 µg/cm2. The study highlights the importance of synergizing UAV and satellite data which enables effective monitoring of canopy chlorophyll in small agricultural lands.

Single-phase-to-ground fault in non-effectively grounded systems poses serious public safety concern. Along with increasing deployment of low-voltage (LV) sensors, new opportunities and ways are created for solving this traditional problem in distribution systems. A new approach, identifying primary single-phase-to-ground line fault and locating it through measurement data from smart sensors deployed in the secondary side of distribution transformers, is proposed in this paper. When a fault occurs in primary distribution feeder, the transient aerial mode voltage (AMV) of each LV smart sensor is calculated. The fault line can be identified by an abrupt change value of AMV at LV side. Next, by finding the position of the maximum value of this AMV from LV sensors on the fault line, the fault location can be pinpointed to two neighboring distribution transformers. The proposed method has been proven to be efficient and reliable by simulation studies.

The performance of a small reconfigurable unmanned aerial vehicle (UAV) is evaluated, 1 combining a multidisciplinary approach in computational analysis of additive manufactured struc- 2 tures, fluid dynamics, and experiments. Reconfigurable UAVs promise cost savings and efficiency 3 without sacrificing performance, while demonstrating versatility to fulfill different military mission 4 profiles. The use of computational fluid dynamics (CFD) in UAV design produces higher accuracy 5 aerodynamic data, which is particularly important for complex aircraft concepts such as blended 6 wing bodies. To address challenges relating to anisotropic materials, the Tsai-Wu failure criterion is 7 applied to structural analysis, using CFD solutions as load inputs. Aerodynamic performance results 8 show the low-speed variant attains an endurance of 1 hour, 48 minutes, whereas for its high-speed 9 counterpart, it is 29 minutes at 66.7% higher cruise speed. Each variant serves different aspects of 10 small UAS deployment in combat, with low-speed envisioned for close surveillance, and high-speed 11 for incursions. Experimental and simulation results suggest room for design iteration, in wing area 12 and geometry adjustments. Structural simulations demonstrated the need for airframe improvements 13 in the low-speed configuration. This paper highlights the potential of reconfigurable UAVs to disrupt 14 the industry, advocating for further research and design improvements.

Aiming at the problem of low charging efficiency caused by the scattered sensor nodes in traditional wireless rechargeable sensor networks (WRSNs), a UAV-assisted WRSN Online Charging Strategy Based on Dynamic Queue and Improved K-means (UOCS) is proposed. The scheme assumes that the energy consumption of nodes is unpredictable, and only generates charging requests when the energy is lower than a threshold, and performs on-demand responses to nodes that issue charging requests. The scheme combines the characteristics of one-to-one charging of UAVs, the selection and allocation timing of waiting queues and the number of UAVs, and the improved K-means partitioning based on space-time coordination(SPKM), which simplifies the problem of coordinated charging of multiple UAVs and maximizes energy. Using the efficiency and charging success rate, the optimal charging trajectory can be found under the constraint that the node will not starve to death due to power shortage. Finally, a simulation comparison experiment is carried out with the existing UAV charging scheduling strategy. UOCS achieves the optimal node survival rate with low algorithm complexity.

Automated machine learning (AutoML) has been heralded as the next wave in artificial intelligence with its promise to deliver high performance end-to-end machine learning pipelines with minimal effort from the user. However, despite AutoML showing great promise for computer vision tasks, to the best of our knowledge, no study has used AutoML for image-based plant phenotyping. To address this gap in knowledge, we examined the application of AutoML for image-based plant phenotyping using wheat lodging assessment with UAV imagery as an example. We compared the performance of an open-source AutoML framework, AutoKeras in image classification and regression tasks to transfer learning using modern convolutional neural network (CNN) architectures. For image classification which classified plot images as lodged or non-lodged, transfer learning with Xception and DenseNet-201 achieved best classification accuracy of 93.2%, whereas Autokeras had 92.4% accuracy. For image regression which predicted lodging scores from plot images, transfer learning with DenseNet-201 had the best performance (R2=0.8303, RMSE=9.55, MAE=7.03, MAPE=12.54%), followed closely by AutoKeras (R2=0.8273, RMSE=10.65, MAE=8.24, MAPE=13.87%). Interestingly, in both tasks, AutoKeras models had up to 40-fold faster inference times compared to the pretrained CNNs. The merits and drawbacks of AutoML compared to transfer learning for image-based plant phenotyping are discussed.

Object segmentation on remotely-sensed images: aerial (or very high resolution, VHS) images and satellite (or high resolution, HR) images, has been applied to many application domains, especially road extraction in which the segmented objects are served as a mandatory layer in geospatial databases. Several attempts in applying deep convolutional neural network (DCNN) to extract roads from remote sensing images have been made; however, the accuracy is still limited. In this paper, we present an enhanced DCNN framework specifically tailored for road extraction on remote sensing images by applying landscape metrics (LMs) and conditional random fields (CRFs). To improve DCNN, a modern activation function, called exponential linear unit (ELU), is employed in our network resulting in a higher number of and yet more accurate extracted roads. To further reduce falsely classified road objects, a solution based on an adoption of LMs is proposed. Finally, to sharpen the extracted roads, a CRF method is added to our framework. The experiments were conducted on Massachusetts road aerial imagery as well as THEOS satellite imagery data sets. The results showed that our proposed framework outperformed Segnet, the state-of-the-art object segmentation technique on any kinds of remote sensing imagery, in most of the cases in terms of precision, recall, and F1.

New technologies have resulted in a better understanding of blood and lymphatic vascular heterogeneity at the cellular and molecular levels. Meanwhile, we still need to learn more about the heterogeneity of the cardiovascular and lymphatic systems between different species at the anatomical and functional levels. Even the deceptively simple question of the functions of fish lymphatic vessels has yet to be conclusively answered. The most common interpretation assumes a similar dual setup of the vasculature in zebrafish and mammals: on the one hand, there is the cardiovascular, circulatory system, and on the other hand, the lymphatic vascular system, in which the unidirectional flow is derived from surplus interstitial fluid and returned into the cardiovascular system. A competing interpretation questions the identity of the lymphatic vessels in fish as at least some of them receive their flow from arteries via specialised anastomoses, neither requiring an interstitial source for the lymphatic flow nor stipulating unidirectionality. In this alternative view, the “fish lymphatics” are a specialised subcompartment of the cardiovascular system, called the “secondary vascular system” (SVS). Many of the contradictions found in the literature appear to stem from the fact that the SVS develops in part or its entirety from an embryonic lymphatic vascular system by transdifferentiation. Future research needs to establish the extent of embryonic transdifferentiation of lymphatics into SVS blood vessels. Similarly, more insight is needed into the molecular regulation of vascular development in fish. Most fish do feature more than the five VEGF genes and three VEGF receptor genes that we know from mice or humans, and the relative tolerance of fish to whole genome and gene duplications could underlie the evolutionary diversification of the vasculature. This review discusses the key elements of the fish lymphatics versus the SVS and attempts to draw a picture coherent with the existing data, including the phylogenetic knowledge.

Since the early 1980s, a large number of research works on enzymes from the red king crab hepatopancreas have been conducted. These studies have been relevant both from a fundamental point of view for studying the enzymes of marine organisms and in terms of the rational management of nature to obtain new and valuable products from the processing of crab fishing waste. Most of these works were performed by Russian scientists due to the area and amount of waste of red king crab processing in Russia (or the Soviet Union). However, the close phylogenetic kinship and the similar ecological niches of commercial crab species and the production scale of the catch provide the bases for the successful transfer of experience in the processing of red king crab hepatopancreas to other commercial crab species mined worldwide. This review describes the value of recycled commercial crab species, discusses processing problems, and suggests possible solutions to these problems. The main emphasis is placed on the enzymes of the hepatopancreas as the most highly salubrious product of waste processed from red king crab fishing.

Urban tree classification is pivotal in urban planning and management, facilitating informed decision-making processes. In this study, we introduce a novel data presentation method termed Pseudo Tree Crown, designed to enhance the accuracy and efficiency of urban tree classification. Leveraging the latest advancements in artificial intelligence (AI), we employ a state-of-the-art classification scheme, PyTorch, to maximize the accuracy of tree classification. Our results demonstrate a robust classification accuracy of over 95% from high spatial resolution imagery from Unmanned Aerial Vehicle (UAV), underscoring our proposed approach’s effectiveness. Moreover, the adaptability of our method renders it applicable to various study areas, highlighting its versatility and potential for widespread implementation in urban planning and management initiatives.

This paper presents the second stage of a tandem fixed wing unmanned aerial vehicle (UAV) aerodynamic development. In the initial stage, the UAV was optimized analyzing its characteristics only in symmetrical flight conditions. Posted requirements were that both wings should produce relevant positive lift, the initial stall must occur on the front wing first, the center of pressure should be close to the center of gravity, while longitudinal static stability should be in the optimum range. Computational fluid dynamics (CFD) analyses were performed, where applied calculation model was derived from the authors’ previous successful projects. The eight version TW V8 has satisfied all longitudinal requirements. Lateral-directional CFD analyses of V8 showed that the ratio of the lateral and directional stability at the nominal cruising regime was optimal, but both lateral and directional static stabilities were too high. On further development versions the lower vertical tail was eliminated, negative dihedral was implemented on the front wing, and four inverted blended winglets were added. Version TW V14 has largely improved lateral and directional stability characteristics, while their optimum ratio at cruising regime was preserved. Longitudinal characteristics were also well preserved. Maximum lift coefficient and lift-to-drag ratio were increased, compared to the V8.

Insulator defect detection is of great significance to compromise the stability of the power transmission line. The state-of-the-art network of object detection, YOLOv5, has been widely used on insulator and defect detection. However, YOLOv5 network has some limitations like poor detection rate and high computational loads in detecting small insulator defects. To solve these problems, we proposed a light-weight network for insulator and defect detection. In this network, we introduced Ghost module into YOLOv5 backbone and neck to reduce the parameters and model size to enhance the performance in unmanned aerial vehicles (UAVs). Besides, we added small object detection anchors and layers for small defect detection. In addition, we optimized the backbone of YOLOv5 by applying convolutional block attention module (CBAM) to focus on critical information for insulator and defect detection and suppress uncritical information. The experiment result shows the mean average precision (mAP) 0.5 and the mAP0.5:0.95 of our model can reach 99.4% and 91.7%, the parameters and model weight are reduced to 3807372 and 8.79M, which can easily deploy to embedded devices like UAVs. And the speed of detection can reach 10.9ms/image, which can meet the real-time detection requirement.

Low-cost miniature hyperspectral and thermal cameras onboard lightweight unmanned aerial vehicles (UAV) bring new opportunities for monitoring land surface variables at unprecedented fine spatial resolution with acceptable accuracy. This research applies hyperspectral and thermal imagery from a drone to quantify upland rice growth and water use efficiency (WUE) after biochar application in a Costa Rican dry region. The field flights were conducted over two experimental groups with bamboo biochar and sugarcane biochar amendments and one control group without biochar application. Rice canopy biophysical variables were estimated by inversion of a canopy radiative transfer model on hyperspectral reflectance. Variations in gross primary production (GPP) and WUE across treatments were estimated from the normalized difference vegetation index (NDVI), canopy chlorophyll content (CCC), and evapotranspiration. We found that GPP was increased by 41.9±3.4 % when using bamboo biochar and 17.5±3.4 % when using sugarcane biochar, which was probably due to higher soil moisture in the biochar-amended plots and led to significantly higher WUE by 40.8±3.5 % in bamboo biochar and 13.4±3.5 % in sugarcane biochar. This study demonstrated the use of hyperspectral and thermal imagery from drone to provide indicators for quantifying biochar effects on tropical dry cropland by integrating with ground point samples and physical models.

Ponderosa pine is an integral part of the forested landscape in the western US; it is the dominant tree species on landscapes that provide critical ecosystem services. Moderate drought tolerance allows it to occupy the transition zone between forests and open woodlands and grasslands. Increases in stand density resulting from wildfire suppression, combined with lengthening, intensifying and more frequent droughts have resulted in reduced tree vigor and stand health in dry ponderosa pine throughout its range. To address a management need for efficient landscape-level surveys of forest health, we used Random Forests to develop an object-oriented classification of individual tree crowns (ITCs) into vigor classes using existing, agency acquired 4-band aerial imagery. Classes of tree vigor were based on quantitative physiological and morphological attributes established in a previous study. We applied our model across a landscape dominated by ponderosa pine with a variety of forest treatments to assess their impacts on tree vigor and stand health. We found that stands that were both thinned and burned had the lowest proportion of low vigor ITCs, and that stands treated before the 2014-2016 drought had lower proportions of low vigor ITCs than stands treated more recently (2016). Upland stands had significantly higher proportions of low vigor trees than lowland stands. Maps identifying the low vigor ITCs would assist managers in identifying priority stands for treatment and marking trees for harvest or retention. These maps can be created using already available imagery and GIS software.

In recent times, the application areas of unmanned aerial vehicles (UAVs) are expanding, and autonomous functions are becoming increasingly prominent, eliminating the need for remote pilot control. Certain applications require the drone to use a docking station autonomously, even for wireless charging. This involves performing autonomous takeoff and landing, where high preci-sion is demanded. These maneuvers can be significantly influenced by environmental factors. The effectiveness of the precision landing function of the DJI Mavic 2 Pro UAV in strong winds is in-vestigated in this article based on experimental measurements conducted outdoors. New findings on the precision landing capabilities of the DJI Mavic 2 Pro UAV are presented in this paper, par-ticularly under windy conditions. The efficiency of precision landing is compared in cases when the precision landing function is enabled versus when it is disabled. It is investigated whether there is a statistically significant difference between the two landing modes in strong winds and it is examined whether the DJI Mavic 2 Pro can perform precision landing on a surface of a specific size in windy conditions. This helps to predict the suitability of the DJI Mavic 2 Pro for precise landing and wireless charging at a docking station in strong winds and high gusts.

This study explores fault-tolerant consensus in leader-following heterogeneous multi-agent systems, focusing on actuator failures in uncrewed aerial vehicles (UAVs) and uncrewed ground vehicles (UGVs). An optimization-based fault-tolerant consensus algorithm is proposed. The algorithm utilizes the Euler-Lagrange formula to ensure system consistency under actuator failures, with the Lyapunov stability theory proving the asymptotic stability of the consistency error. The algorithm is applied to heterogeneous multi-agent systems of UAVs and UGVs to derive optimal fault-tolerant consensus control laws for each vehicle type. Simulation experiments give evidence for the feasibility of the proposed control strategy.

Recently, methods to detect DoS and spoofing attacks that occur on in-vehicle networks using CAN Protocol are being studied through deep learning models such as CNN, RNN, and LSTM. These studies have produced significant results in the field of In-Vehicle Network attack detection using deep learning models. In addition, significant results are being achieved through research on applying time series-based deep learning models such as LSTM to detect DoS attacks and replay attacks occurring in in-drone networks by expanding them to drones using the UAVCAN protocol. In this paper, we conducted an experiment to detect in-drone network attacks through non-time series analysis using machine learning models and deep learning models, and through appropriate learning for each attack type, it can also be analyzed through non-time series analysis. The results showed that it was possible to detect attacks.

A novel technique suitable for propulsion of small unmanned aerial vehicles (UAV) is discussed in this paper. This approach utilizes the rotational energy of airborne gyro rotors and converts it into translational propulsion for the vehicle. The energy conversion is achieved by generating precisely directed centrifugal force pulses through short-duration rotor unbalances. The accurate control of the timing and magnitude of these unbalances is crucial for successful propulsion generation. Our first-order theory of controlled unbalance propulsion (CUP) predicts the potential for achieving high translational accelerations and vehicle velocities up to orbital levels. Power saving levitation of UAVs can be attained. In this paper, we provide traceable evidence that pulsed centrifugal propulsion is based on well-established laws of physics and can be realized using state-of-the-art technologies.

Tail-sitters aim to combine the advantages of fixed-wing and rotor-craft, but demand a robust and fast stabilization strategy to perform vertical maneuvers and transitions to and from aerodynamic flight. The research conducted in this work intends to assess the performance of nonlinear control strategies to stabilize the attitude of a X-Vert VTOL aircraft when hovering, comparing existing solutions to applications of Nonlinear Dynamics Inversion (NDI) and its incremental version, INDI. Such controllers are implemented and tuned in simulation in order to stabilize a model of the tail-sitter , complemented by estimation methods that allow to feed back the necessary variables. These estimators and controllers are then implemented in a microcontroller, validating them in a Hardware-in-the-Loop (HITL) scenario, with simple maneuvers in vertical flight. Lastly, the developed control solutions are used to stabilize the aircraft in experimental flight, being monitored by a motion capture system. The experimental results allow to validate the model of the X-Vert and compare the effectiveness of the different control solutions in stabilizing it, with the INDI presenting itself as a more robust control strategy, with better tracking capabilities and less actuator demands.

The urban heat island (UHI) and its intensity is one of the phenomena that are of determining importance for the comfort of living in cities and their sustainable development in the face of deepening climate change. The study is objectively difficult due to the large dynamics like land cover and the considerable diversity of land use patterns in urban areas. Various approaches and methods are used in research practice, based both on ground measurements and in the use of remote methods and satellite information. However, most of these approaches provide information with problematic spatial and temporal resolution, making them difficult to apply for sustainable urban planning purposes, which require adequate information provision. This paper presents the results of the application of thermal photogrammetry based on the use of unmanned aerial systems (UAS), combined with geographic information systems (GIS), in the study of surface urban heat island intensity (SUHI), at the local level for the largest housing complex in Bulgaria - Lyulin district of the capital of Sofia city. The studies were carried out during a heat wave in July 2023. A difference of 16.5oC was found between locations with SUHI occurrence and suburban areas. The information benefits of locally addressed data and their direct applicability are discussed to support decision-making processes in the planning and management of urban areas, including their climate adaptation and sustainable development.

The use of unmanned aerial aircrafts (UAVs) is governed by strict regulatory frameworks that prioritize safety. To guarantee safety, it is necessary to acquire and maintain situational awareness (SA) throughout the operation. Existing Canadian regulations require pilots to operate their aircrafts in visual line-of-sight. Therefore, the task of acquiring and maintaining SA primary falls to the pilots. However, the development of aerial transport is entering a new era with the adoption of a highly dynamic and complex system known as advanced air mobility (AAM), which involves UAV operating autonomously beyond visual line-of-sight. SA must therefore be acquired and maintained primarily by each UAV through specific technologies and procedures. In this paper, we review these technologies and procedures in order to decompose the SA of the UAV in the AAM. We then use the system modeling language to provide a high-level structural and behavioral representation of the AAM as a system having UAV as its main entity. In a case study, we analyze one of the flagships UAVs of our industrial partner. Results show that this UAV doesn’t have all the technologies and methodologies necessary to achieve all the identified SA goals identified for the safety of the AAM. This work is a theoretical framework intended to contribute to the realization of the AAM, and we also expect to impact the future design and utilization of UAVs.

In this paper, we address the problem of car detection from aerial images using Convolutional Neural Networks (CNN). This problem presents additional challenges as compared to car (or any object) detection from ground images because features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of two state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, and YOLOv3, which is known to be the fastest detection algorithm. We analyze two datasets with different characteristics to check the impact of various factors, such as UAV's altitude, camera resolution, and object size. The objective of this work is to conduct a robust comparison between these two cutting-edge algorithms. By using a variety of metrics, we show that YOLOv3 yields better performance in most configurations, except that it exhibits a lower recall and less confident detections when object sizes and scales in the testing dataset differ largely from those in the training dataset.

In this paper, we address the problem of car detection from aerial images using Convolutional Neural Networks (CNN). This problem presents additional challenges as compared to car (or any object) detection from ground images because features of vehicles from aerial images are more difficult to discern. To investigate this issue, we assess the performance of two state-of-the-art CNN algorithms, namely Faster R-CNN, which is the most popular region-based algorithm, and YOLOv3, which is known to be the fastest detection algorithm. We analyze two datasets with different characteristics to check the impact of various factors, such as UAV’s altitude, camera resolution, and object size. The objective of this work is to conduct a robust comparison between these two cutting-edge algorithms. By using a variety of metrics, we show that none of the two algorithms outperforms the other in all cases.

Human detection from Unmanned Aerial Vehicles (UAV) is gaining popularity in the field of disaster management, crowd counting, people monitoring. Real time human detection from UAV is a challenging task, because of many constraints involved. This study proposes a system for real time detection of humans on videos captured from UAVs addressing three of these constraints namely, flying height, computation time and scale of viewing. The proposed method integrated an android application with a binary classifier based on Haar-features to automatically detect human / non-human class from UAV images. The video frames were parsed and detected humans from image frames were geo-localized and visualized on Google Earth. The performance was evaluated for geo-localization accuracy, computation time and detection accuracy, considering human coverage – pixel size relationship for various heights and scale factor. Based on flying height - human size relationship and tradeoff between detection accuracy vs computation time, the study came up with optimal parameters for OpenCV’s cv2.cascadeClassifier. detectMultiScale function. This paper establishes a strong ground for further research relating to real time human detection from UAV.

Drone imagery has the potential to enrich urban planning and preservation, especially where it converges with the growing creation and use of 3D models. The authors have conducted a systematic literature review of articles published between 2002 and 2022, drawing from reputable academic repositories including Science Direct, Web of Science, and China National Knowledge Infrastructure (CNKI), to identify current gaps in existing research in the application of UAVs to the creation of 3D models in urban planning and preservation. Findings indicate five research shortcomings: limited participation of planning experts, research focus imbalance, lack of usage for special scenarios, lack of integration with smart city planning, and limited interdisciplinary collaboration. The study also acknowledges current limitations with UAV applications and discusses possible countermeasures as well as future prospects.

Drone and aerial remote sensing images are widely used, but their imaging environment is complex and prone to image blurring. Existing CNN deblurring algorithms usually use multi-scale fusion to extract features in order to make full use of aerial remote sensing blurred image information, but images with different degrees of blurring use the same weights, leading to increasing errors in the feature fusion process layer by layer. Based on the physical properties of image blurring, this paper proposes an adaptive multi-scale fusion blind deblurred generative adversarial network (AMD-GAN), which innovatively applies the degree of image blurring to guide the adjustment of the weights of multi-scale fusion, effectively suppressing the errors in the multi-scale fusion process and enhancing the interpretability of the feature layer. The research work in this paper reveals the necessity and effectiveness of a priori information on image blurring levels in image deblurring tasks. By studying and exploring the image blurring levels, the network model focuses more on the basic physical features of image blurring. Meanwhile, this paper proposes an image blurring degree description model, which can effectively represent the blurring degree of aerial remote sensing images. The comparison experiments show that the algorithm in this paper can effectively recover images with different degrees of blur, obtain high-quality images with clear texture details, outperform the comparison algorithm in both qualitative and quantitative evaluation, and can effectively improve the object detection performance of aerial remote sensing blurred images. Moreover, the average PSNR of this paper's algorithm tested on the publicly available dataset RealBlur-R reached 41.02dB, surpassing the latest SOTA algorithm.

Unmanned aerial vehicles (UAVs) represent a new model of social robots for home care of dependent persons. In this regard, this article introduces a study on people’s feeling of safety and comfort while watching the monitoring trajectory of a quadrotor dedicated to determining their condition. Three main parameters are evaluated: the relative monitoring altitude, the monitoring velocity and the shape of the monitoring path around the person (ellipsoidal or circular). For this purpose, a new trajectory generator based on a state machine, which is successfully implemented and simulated in MATLAB/Simulink®, is described. The study is carried out with 37 participants using a virtual reality (VR) platform based on two modules, UAV Simulator and VR Visualiser, both communicating through the MQTT protocol. The participants’ preferences have been a high relative monitoring altitude, a high monitoring velocity and a circular path. These choices are a starting point for the design of trustworthy socially assistive UAVs flying in real homes.

In scenarios such as natural disasters and military strike, it is common for unmanned aerial vehicles (UAVs) to form groups to execute reconnaissance and surveillance. To ensure the effectiveness of UAV communications, repeated resource acquisition issues and transmission mechanism design need to be addressed urgently. In this paper, we build an information interaction scenario in a Flying Ad-hoc network (FANET). The data transmission problem with the goal of throughput maximization is modeled as a coalition game framework. Then, a novel mechanism of coalition selection and data transmission based on group-buying is investigated. Since large-scale UAVs will generate high transmission overhead due to the overlapping resource requirements, we propose a resource allocation optimization method based on distributed data content. Comparing existing works, a data transmission and coalition formation mechanism is designed. Then the system model is classified into graph game and coalition formation game. Through the design of the utility function, we prove that both games have stable solutions. We also prove the convergence of the proposed approach with coalition order and Pareto order. Binary log-linear learning based coalition selection algorithm (BLL-CSA) is proposed to explore the stable coalition partition of system model. Simulation results show that the proposed data transmission and coalition formation mechanism can achieve higher data throughput than the other contrast algorithms.

The Third U.S. National Marine Environmental DNA Workshop on June 2–5, 2024 brought together researchers, practitioners, and policymakers to overview and discuss: (1) environmental (e)DNA – defined as “DNA in the environment” – as a strategic U.S. national priority, (2) eDNA tool readiness and approaches to support decision-making, (3) emerging eDNA technologies, and (4) plans for implementation and adoption for environmental monitoring and ecosystem assessment “from microbes through whales”. Uses and applications of eDNA have been revolutionizing exploration, measurement, and monitoring of biodiversity across ecosystems, including the marine and freshwater aquatic environments emphasized in the Workshop, as well as terrestrial and aerial systems. The Workshop featured the launch of the new U.S. National Aquatic Environmental DNA Strategy from the White House Office of Science and Technology Policy (2024), with aquatic defined as freshwater to ocean waters, across all salinities; the Strategy was developed using outputs and a taskforce from the 2022 Second Workshop. The 2.5-day Third Workshop included speaker presentations, panel discussions, and networking on eDNA tools, applications, use, upscaling, and their future. The Workshop led into Capitol Hill Ocean Week (CHOW) events (June 5–6, 2024), during World Ocean Week. Note that this summary paper represents the author’s take-away view and interpretation, and not necessarily the overall views or consensus of the U.S. Government, Smithsonian Institution, workshop organizers, or speakers.

With Unmanned Aerial Vehicles (UAV), a swift response to urgent needs like search \& rescue missions or medical deliveries can be realized. Simultaneously, the legislator is establishing so-called geographical zones, which restrict UAV operations to mitigate the air and ground risk to third parties. These geographical zones serve a particular safety interest, but they may also hinder the efficient usage of UAVs on time-critical missions with a range-limiting battery capacity. In this study, we address a facility location problem for up to two UAV hangars with a robust optimization model considering demand hotspots, geographical zones as restricted areas, a standard mission to satisfy battery capacity constraints, and the impact of wind scenarios. To this end, water rescue missions are used exemplary, for which positive and negative location factors for UAV hangars and areas of increased drowning risk as demand points are derived from open-source georeferenced data. Optimal UAV mission trajectories are computed with an A* algorithm considering five different restriction scenarios. As this pathfinding is very time-consuming, binary occupancy grids and image processing algorithms accelerate the computation by identifying either entirely inaccessible or restriction-free connections beforehand. For the optimal UAV hangar locations, we maximize accessibility while minimizing the service time to the hotspots, resulting in a decrease from the average service time of 570.4 s for all facility candidates to 351.1 s for one and 287.2 s for two optimal UAV hangar locations.

This paper presents a novel approach for accurate counting and localization of tropical plants in aerial images that is able to work in new visual domains in which the available data is not labeled. Our approach uses deep learning and domain adaptation, designed to handle domain shift between the training and test data, which is a common challenge in this agricultural applications. This method uses a source dataset with annotated plants and a target dataset without annotations, and adapts a model trained on the source dataset to the target dataset using unsupervised domain alignment and pseudolabeling. The experimental results show the effectiveness of this approach for plant counting in aerial images of pineapples under significative domain shift, achieving a reduction up to 97% in the counting error when compared to the supervised baseline.

This manuscript explores unmanned aerial vehicle DC motor control performance efficacy of deterministic artificial intelligence in comparison to model-following adaptation, particularly a direct self-tuner with filtering. The deterministic artificial intelligence model made use of self-awareness statements to overcome error in response to permutations of the multi-duty cycle square wave that served as the system input. It can be seen that (despite equivalently powerful estimation techniques) deterministic artificial intelligence provided far superior results: a reduction in peak initial transient error of 55%, and a mean error reduction over 81% with over 65% reduction in error standard deviation compared to a state-of-the-art nonlinear adaptive control method. Deterministic artificial intelligence also was able to very closely track at the switching of the input control, while the benchmark nonlinear adaptive control failed to respond as quickly at these points.

Detecting, tracking, and classifying unmanned aerial vehicles (UAVs) in a swarm presents significant challenges due to their small and diverse radar cross-sections, multiple flight altitudes, velocities, and close trajectories. To overcome these challenges, adjustments of the radar parameters and/or position of the radar (for airborne platforms) are often required during runtime. The runtime adjustments help to overcome the anomalies in the detection, tracking, and classification of UAVs. The runtime adjustments are performed either manually or through fixed algorithms, each of which can have its limitations for complex and dynamic scenarios. In this work, we propose the use of multi-agent reinforcement learning (RL) to carry out the runtime adjustment of the radar parameters and position of the radar platform. The radar used in our work is a multibeam multifunction phased array radar (MMPAR) placed onboard UAVs. The simulations show the cognitive adjustment of the MMPAR parameters and position of the airborne platform using RL helps to overcome anomalies in the detection, tracking, and classification of UAVs in a swarm. A comparison with other artificial intelligence (AI) algorithms shows that RL performs better due to runtime learning of the environment through rewards.

This work is part of an ongoing research project to develop an unmanned flying social robot to monitor dependants at home in order to detect the person’s state and bring the necessary assistance. In this sense, this paper focuses on the description of a virtual reality (VR) simulation platform for the monitoring process of an avatar in a virtual home by a rotatory-wing autonomous unmanned aerial vehicle (UAV). This platform is based on a distributed architecture composed of three modules communicated through the Message Queue Telemetry Transport (MQTT) protocol: the UAV Simulator implemented in MATLAB/Simulink, the VR Visualiser developed in Unity, and the new emotion recognition (ER) System developed in Python. Using a face detection algorithm and a convolutional neural network (CNN), the ER System is able to detect the person’s face in the image captured by the UAV’s on-board camera and classify the emotion among seven possible ones (surprise, fear, happiness, sadness, disgust, anger or neutral expression). The experimental results demonstrate the correct integration of this new computer vision module within the VR platform, as well as the good performance of the designed CNN, with around 85% in the F1-score, a mean of the precision and recall of the model. The developed emotion detection system can be used in the future implementation of the assistance UAV that monitors dependent people in a real environment, since the methodology used is valid for images of real people.

Debris-covered glaciers are a notable feature in the greater Himalaya, and their ongoing mass loss under changing climate will affect the water resources of over a billion people. The current knowledge of the mass balance of Himalayan glaciers is restricted by the paucity of in-situ measurements of glaciers in both space and time, as well as the resolution of satellite remote sensing imageries. Recently, the use of Unmanned Aerial System (UAS) imagery has shown the potential to bridge this gap by enabling very detailed monitoring of inaccessible glacial areas. UAS imagery-based monitoring of Himalayan glaciers has so far been limited to a single glacier in the entire Himalaya, providing a limited understanding of spatial variability in glacier mass balance and driving factors. In the first UAS based glacial mass change estimation in the trans-Himalaya, we conducted two Unmanned Aerial System (UAS) surveys (May and November 2019) over the debris-covered Annapurna III glacier in the Himalaya. We performed Structure-from-Motion (SfM) analysis and utilized differential GPS field observations to derive geometrically accurate point clouds, ortho-mosaics and digital surface models (DSMs). The glacial volumetric loss was estimated from DSM differencing, and the magnitude and spatial variability of glacier surface change was derived from 3-D differencing of point clouds. Results revealed a heterogeneous glacial melt pattern, with an average elevation loss of 0.89 m during the monitored time period. The majority of the glacial tongue exhibited surface lowering except the area above and around the glacial snout that surprisingly exhibited significant elevation gain. Both the highest magnitude of mass loss and the highest spatial variability in mass change was observed in areas with exposed ice-cliffs and supraglacial ponds. Glacial surface velocity derived from manual feature tracking showed velocity ranging from 0-4.1 m. A detailed evaluation of specific areas allowed an improved understanding of the complex interplay of factors leading to observed surface change. Our findings expand the extent of UAS based monitoring of debris-covered glaciers in the Himalaya and conclude that UAS derived 3D topographic products will become increasingly important for monitoring of thinning debris-covered glaciers.

Structure from Motion with Multi-View Stereo photogrammetry (SfM) is increasingly utilised in geoscience investigations as a cost-effective method of acquiring high resolution (sub-meter) topographic data, but has not been thoroughly tested in gullied savanna systems. The aim of this study was to test the accuracy of topographic models derived from aerial (via an Unmanned Aerial Vehicle, ‘UAV’) and ground-based (via a handheld digital camera, ‘Ground’) SfM in modelling a hillslope gully system in dry-tropical savanna, and to assess the strengths and limitations of the approach at different scales. A UAV survey covered an entire hillslope gully system (0.715 km²), whereas a Ground survey covered a single gully within the broader system (650 m²). SfM topographic models, including Digital Surface Models (DSM) and dense point clouds, were compared against RTK-GPS point data and a pre-existing airborne LiDAR Digital Elevation Model (DEM). Results indicate UAV SfM can deliver topographic models with a resolution and accuracy suitable to define gully systems at a hillslope scale (e.g., 0.1 m resolution with ~ 0.5 – 1.3 m elevation error), while ground-based SfM is more capable of quantifying gully morphology (e.g., 0.01 m resolution with ~ 0.1 m elevation error). Key strengths of SfM for these applications include: the production of high resolution 3D topographic models and ortho-photo mosaics, low survey instrument costs (< $AUD 3,000); and rapid survey time (4 and 2 hours for UAV and Ground survey respectively). Current limitations of SfM include: difficulties in reconstructing vegetated surfaces; uncertainty as to optimal survey and processing designs; and high computational demands. Overall, this study has demonstrated great potential for SfM to be used as a cost-effective tool to aid in the mapping, modelling and management of hillslope gully systems at different scales, in tropical savanna landscapes and elsewhere.

TThe following three objectives were tested in this study: (1) investigate the utility of low-altitude remote sensing using UAS technology to compare the effects of different N application systems in rice production; (2) use spatial extrapolation to scale up plot-level generated to farmer field rice yield data based on crop spectral signatures, and (3) predict and map out rice productivity as a function of N placement systems. Images were captured on a UAV platform at midseason of the rice crop. Orthomosaics were developed for selected fields in rice-producing zones. Grain yields were assessed from low, medium, and high crop health plots delineated based on NDVI values. On the plot scale, UDP outyielded non-UDP by 0.84%. Individual plot yield data were scaled up to the farmer field level through Jenks natural breaks classification and es-tablishing an empirical relationship between OSAVI and plot yields. Assessment of the scaled-up field levelfield-level data also confirmed the superiority of UDP N man-agement over the non-UDP systems in promoting rice yields. Scaling up plot scale da-ta to whole field levels also facilitated generating and mapping expected yield maps for individual farmer fields in the three zones studied. This study has established a tangible simple but tangible protocol protocol for predicting and mapping rice yields in small-scale farmer fields using UAS data.

In recent years, pedestrian re-identification has gained a lot of interest due to its importance for many purposes such as security and safety. Many types of solutions have been proposed to solve this problem, where the majority are based on a features extraction such Convolution Neural Networks (CNNs). These approaches assume that only the identities that are in the training data can be recognized. The pedestrians in the training data are called In distribution (ID). However, in real world scenarios, new pedestrians and objects can appear in the scene and the model should detect them as Out Of Distribution (OOD). In our previous study , we proposed a pedestrian re-identification based on von-Mises Fisher (vMF) distribution. Each identity is embedded in the unit sphere as a compact vMF distribution far from other identities distributions. The embedding is done through a base CNN. Recently, proposed a framework called Virtual Outlier Synthetic (VOS), that detects OOD based on synthesising virtual outlier in the embedding space in an online manner. Their approach assumes that the samples from the same object maps to a compact space. This assumption aligns with the vMF based approach. Therefore, in this paper, we revisited the vMF approach and merged with VOS in order to detect OOD data points. We present our approach to merge both frameworks. We conducted several experiments to evaluate our proposed framework. Results showed that our framework was able to detect new pedestrian that do not exist in the training data in the inference phase. It also slightly helped to improve the re-identification performance.

The location and number data of individual fruit trees are critical for planting area investigation, fruit yield prediction, and smart orchard management and planning. These data are conventionally obtained through manual investigation and statistics with time-consuming and laborious effort. Object detection models in deep learning used widely in computer vision could provide an opportunity for accurate detection of individual fruit trees, which is essential for rapidly obtaining the data and reducing human operations errors. This study proposes an approach to detecting individual fruit trees and mapping their spatial distribution by integrating deep learning with unmanned aerial vehicle (UAV) remote sensing. UAV remote sensing collected high-resolution true-color images of fruit trees in the experimental pomelo tree orchards in Meizhou city, South China. An image dataset of deep learning samples of individual pomelo trees (IPTs) was constructed through visual interpretation and field investigation based on the fruit tree images captured by UAV remote sensing. Four different scales of YOLOv5 (YOLOv5s, YOLOv5m, YOLOv5l, and YOLOv5x) for object detection were selected to train, validate, and test on the image dataset of pomelo trees. The results show that the average precision ([email protected]) of the four YOLOv5 models for validation reach 87.8%, 88.5%, 89.1%, and 90.7%, respectively. The larger the model scale, the higher the average accuracy of the detection result of validation. It suggests that YOLOv5x is a preferred high-accuracy model among the YOLOv5 family and is suitable to realize the detection of IPTs. The number of the IPTs in the study area was counted using YOLOv5x, and their spatial distribution map was made using the non-maximum suppression method and ArcGIS software. This study will provide primary data and technical support for smart orchard management in Meizhou city and other fruit-producing areas.

This study introduces a remote sensing application using satellite imagery to survey a network-scale aggregate stockpile inventory. First, a real scale aggregate quarry site was surveyed using a small Unmanned Aerial Vehicle (sUAV) to produce digital terrain models that enabled analysis of aggregate pile geometry. Second, a lab experiment was designed and performed to validate the applicability of close-range Structure from Motion (SfM) photogrammetry for measuring aggregate piles' physical properties such as volume and density. The other part of the lab experiment delved into direct measurement of aggregate density under varying compaction efforts. These experimental results, in conjunction with some simplifying assumptions, enabled the calculation of aggregate stockpile volumes and estimated weights from satellite imagery. We estimated that an inventory of 4.4 and 1.1 million metric tons of crushed aggregates and Reclaimed Asphalt Pavement (RAP), respectively, stockpiled in Washington State for asphalt production in 2017. The merit of producing such database was further showcased in an example on the economic and environmental impacts of material transportation. We approximated that hauling aggregates from quarry plants to construction sites within Washington State incurs a cost of about $50 thousand to over $4 million, consumes about 0.25 to 20 TJ of energy, and emits 20 to over 1,500 tons of CO₂-eq per asphalt plant annually.

Forest canopy gaps are important for the ecosystem dynamics. Depending on tree species, small canopy openings might be also associated to intra-crown porosity and to space between crowns. Yet, little is known on the relationships between the fine-scaled pattern of canopy openings and biodiversity features. This research explored the possibility of i)- mapping forest canopy gaps from a very high resolution orthomosaic (10 cm), processed from a versatile imaging platform such as unmanned aerial vehicles (UAV), ii)- to derive patch metrics that can be tested as covariates of variables of interest for forest biodiversity monitoring. This is attempted in a test area of 240 ha covered by temperate deciduous forest types in Central Italy and containing 50 forest inventory plots of about 530 m². Correlation and linear regression techniques were used to explore relationships between patch metrics and understorey (density, development and species diversity) or forest habitat biodiversity variables (density of micro-habitat bearing trees, vertical species profile, tree species diversity). The results revealed that small openings in the canopy cover (75% smaller than 7 m²) can be faithfully extracted from UAV RGB imagery, using the red band and contrast split segmentation. Highest correlations were observed in the mixed forest (beech and turkey oak), while beech forest had the poorest ones and turkey oak forest displayed intermediate results. Moderate to strong linear relationships were found between gap metrics and understorey variables in mixed forest type, with adjusted R² from linear regression ranging from 0.52 to 0.87. Equally good results, in the same forest types, were observed for forest habitat biodiversity variables (0.52<adjusted R²<0.79) with highest values found for density of trees with microhabitats and vertical species profile. In conclusion, this research highlights that UAV remote sensing can potentially provide covariate surfaces of variables of interest for forest biodiversity monitoring, conventionally collected in forest inventory plots. By integrating the two sources of data, these variables can be mapped over small forest areas with satisfactory levels of accuracy, at a much higher spatial resolution than would be possible by field-based forest inventory solely.

This literature review aims to gain a deeper understanding of Spinosaurus aegyptiacus’s dorsal sail regarding its function as thermoregulation, display, or hydrodynamic. The function of S. aegyptiacus’s dorsal sail has been widely debated for many years. The three most popular hypotheses are that it was used as a biological mechanism of thermoregulation, display, or hydrodynamic. The display hypothesis includes attracting mates, intimidating rivals, and social communication with other members. This literature review delves into each of the three hypotheses using past studies and excavation sites to analyze their validity. For the hydrodynamic aspect, the lifestyle of S. aegyptiacus must first be established. Based on existing research, we cannot conclude that there is a definitive primary function of Spinosaurus aegyptiacus’s dorsal sail. However, it is likely that its sail served the crucial function of display, since it would have a large surface area and could thus be a distinguishable characteristic of S. aegyptiacus that could determine mating success. Additionally, it is probable that the sail had supplementary roles in thermoregulation and hydrodynamics.

Drones, as mobile media of the present day, increase the operational and narrative capabilities of television and accelerate the logistics of shooting. Unmanned aerial vehicles with a camera properly steered by a pilot are able, to some extent, to replace a jimmy jib/crane and a dolly; basic technical devices, used in the studios, enabling the creation of narrative systems of pictures in film and television. Television is more and more often using drone footage to report events, broadcast live, as well as create coverage and television documentaries. In many productions the pilot of the drone simultaneously acts as the drone camera operator, which can improve the effectiveness of shooting, but also carries some risk related to flight safety. The article describes and presents in the form of processed footage the real conditional ties of a Visual Line of Sight (VLOS) flight faced by pilots filming with a drone. VLOS is a type of air operation, which requires maintaining eye contact with the drone. In many countries a drone visibility flight is legally sanctioned as VLOS Operation. An experiment was conducted to investigate the interactions between a human and a machine in airspace steered using a controller with a touchscreen. The drone pilot was considered an integral part of the drone’s flight system control Experimental data was collected with the use of a mobile eye-tracker, video cameras, surveys and pilot declarations. During the experiment, eight television drone pilot operators recaptured a model shot under the regime of VLOS flight at low altitude. They all show that both advanced and beginner pilots did not look at the UAV for over half the time of shot execution. The experiment allowed establishing two coefficients related to the effectiveness of a VLOS flight aiming at filming from the drone. The results point to clear differences in screen perception styles used by drone television pilots.. The coefficients were described in the form of mathematical formulas and their limit values were determined. The research also determines the limits of pilots’ perception, within which they are able to film with a drone. The outcomes may help to optimize the process of aerial filming with the use of a drone, carried out for television, film and other media, as well as in a simulation of such a flight for research and training. From the perspective of media science and social communication, the presented study included a technological component that can be accessed through information science, using statistical models and variable distributions. Media scholars can study the impact of the media without having to look into the metaphorical black box. Computer science opens up this possibility.

In recent years, demand has been increasing for target detection and tracking from aerial imagery via drones using onboard powered sensors and devices. We propose a very effective method for this application based on a deep learning framework. A state-of-art embedded hardware system empowers small flying robots to carry out the real-time onboard computation necessary for object tracking. Two types of embedded modules were developed: one is designed using a Jetson TX or AGX Xavier, and the other is based on an Intel Neural Compute Stick. These are suitable for real-time onboard computing power on small flying drones with limited space. A comparative analysis of current state-of-art deep-learning-based multi-object detection algorithms was carried out utilizing the designated GPU-based embedded computing modules to obtain detailed metric data about frame rates as well as the computation power. We also introduce an effective target tracking approach for moving objects. The algorithm for tracking moving objects is based on the extension of simple online and real-time tracking. It was developed by integrating a deep-learning-based association metric approach (Deep SORT), which uses a hypothesis tracking methodology with Kalman filtering and a deep-learning-based association metric. In addition, a guidance system that tracks the target position using a GPU-based algorithm is introduced. Finally, we demonstrate the effectiveness of the proposed algorithms by real-time experiments with a small multi-rotor drone.

Incremental control strategies provide undeniable advantages for controlling Uncrewed Aerial Vehicles (UAVs) due to their reduced model dependency and accurate tracking capacities. This is of particular relevance for hybrid UAVs, specially for tail-sitters, which perform complex manoeuvres when transitioning to and from aerodynamic flight. This paper summarizes the research undertaken in the application of the Incremental Nonlinear Dynamics Inversion (INDI) and Incremental Backstepping (IBKS) to a small tail-sitter UAV. After the application of these control strategies in a simulated scenario, the implementation in a Micro-Controller Unit was validated in a Hardware-in-the-Loop scenario. Following such validation, indoor experimental trials were undertaken in order to evaluate the capacity of these controllers to stabilize and control the real aircraft. The stabilization in the flight trials was successfully implemented and allow to conclude that both INDI and IBKS are control strategies suited to control a tail-sitter in vertical flight, providing accurate tracking capabilities under smooth actuation. Furthermore, it was found that the IBKS is significantly more computationally demanding than the INDI, although having a global proof of stability that is of interest in aircraft control.

In recent years, the unmanned aerial vehicle (UAV) synthetic aperture radar (SAR) has become a highly sought-after topic for its wide applications in the field of target recognition, detection, and tracking. However, SAR automatic target recognition (ATR) models based on deep neural networks (DNN) are suffering from adversarial examples. Generally, non-cooperators rarely disclose any information about SAR-ATR models, making adversarial attacks challenging. In this situation, we propose Transferable Adversarial Network (TAN) to attack these models with highly transferable adversarial examples. The proposed method improves the transferability via a two-player game, in which we simultaneously train two encoder-decoder models: a generator that crafts malicious samples through a one-step forward mapping from original data, and an attenuator that weakens the effectiveness of malicious samples by capturing the most harmful deformations. In particular, compared to traditional iterative methods, our approach is able to one-step map original samples to adversarial examples, thus enabling real-time attacks. Experimental results indicate that the proposed approach achieves state-of-the-art transferability with acceptable adversarial perturbations and minimum time costs compared to existing attack methods, i.e., it excellently realizes real-time transferable adversarial attacks.

Change Detection (CD) of aerial images refers to identifying and analyzing changes between two or more aerial images of the same location taken at different times. The CD is a highly challenging task due to the need to distinguish relevant changes, such as urban expansion, deforestation, or post-disaster damage assessment, from irrelevant ones, such as light conditions, shadows, and seasonal variations. Many CD papers have recently been published, where most of the papers that proposed a new model contained a comparison between their proposed and state-of-the-art (SOTA) models. While many recent studies propose new deep learning (DL) models for improving CD performance, their comparative analyses are often restricted, lacking comprehensive insights into the proposed models' real-world generalizability, robustness, and performance trade-offs across diverse change characteristics. This paper presents a novel generic framework to systematically benchmark and assess DL-based CD models through three parallel pipelines: 1) cross-testing models on diverse benchmark datasets to evaluate generalization, 2) robustness analysis against different image corruptions, and 3) multi-faceted contour-level analytics evaluating detection sensitivity to change size/complexity. The framework is applied to comparatively evaluate five state-of-the-art DL-based CD models - Changeformer, BIT, Tiny, SNUNet, and CSA-CDGAN. Extensive experiments unveil each model's strengths, limitations and biases, highlighting their relative proficiencies in generalizing across data distributions, resilience to noise corruption, and discriminative capabilities for changes of varying characteristics. The proposed benchmarking framework demonstrates significant potential for guiding the selection of suitable CD models tailored to specific application requirements by comprehensively evaluating their generalizability, robustness, and detection capabilities across diverse real-world scenarios. This systematic evaluation approach can drive future research into developing more robust and versatile CD solutions aligned with practical needs.