Search

Monitoring vast and remote areas like forests using Wireless Sensor Networks (WSNs) presents significant challenges, such as limited energy resources and signal attenuation over long distances due to natural obstacles. Traditional solutions often require extensive infrastructure, which is impractical in such environments. To address these limitations, we introduce the "FloatingBlue" architecture. This architecture, known for its superior energy efficiency, combines Bluetooth Low Energy (BLE) technology with Delay Tolerant Networks (DTN) and data mules. It leverages BLE's low power consumption for energy-efficient sensor broadcasts while utilizing DTN-enabled data mules to collect data from dispersed sensors without constant network connectivity. Deployed in a remote agricultural area in the Amazon region, "FloatingBlue" demonstrated significant improvements in energy efficiency and communication range, with a high Packet Delivery Ratio (PDR). The developed BLE beacon sensor achieved state-of-the-art energy consumption levels, using only 2.25 µJ in sleep mode and 11.8 µJ in transmission mode. Our results highlight "FloatingBlue" as a robust, low-power solution for remote monitoring in challenging environments, offering an energy-efficient and scalable alternative to traditional WSN approaches. [ABSTRACT FROM AUTHOR]

The distribution of tiny sensors over a specific area was first proposed in the late 1990s as a concept known as smart dust. Several efforts focused primarily on computing and networking capabilities, but quickly ran into problems related to power supply, cost, data transmission, and environmental pollution. To overcome these limitations, we propose using paper-based (confetti-like) chemosensors that exploit the inherent selectivity of chemical reagents, such as colorimetric indicators. In this work, cheap and biodegradable passive sensors made from cellulose could successfully indicate the presence of hazardous chemicals, e.g., strong acids, by a significant color change. A conventional color digital camera attached to a drone could easily detect this from a safe distance. The collected data were processed to define the hazardous area. Our work presents a combination of the smart dust concept, chemosensing, paper-based sensor technology, and low-cost drones for flexible, sensitive, economical, and rapid detection of hazardous chemicals in high-risk scenarios. [ABSTRACT FROM AUTHOR]

Single-tree segmentation on multispectral UAV images shows significant potential for effective forest management such as automating forest inventories or detecting damage and diseases when using an additional classifier. We propose an automated workflow for segmentation on high-resolution data and provide our trained models in a Toolbox for ArcGIS Pro on our GitHub repository for other researchers. The database used for this study consists of multispectral UAV data (RGB, NIR and red edge bands) of a forest area in Germany consisting of a mix of tree species consisting of five deciduous trees and three conifer tree species in the matured closed canopy stage at approximately 90 years. Information of NIR and Red Edge bands are evaluated for tree segmentation using different vegetation indices (VIs) in comparison to only using RGB information. We trained Faster R-CNN, Mask R-CNN, TensorMask and SAM in several experiments and evaluated model performance on different data combinations. All models with the exception of SAM show good performance on our test data with the Faster R-CNN model trained on the red and green bands and the Normalized Difference Red Edge Index (NDRE) achieving best results with an F1-Score of 83.5% and an Intersection over Union of 65.3% on highly detailed labels. All models are provided in our TreeSeg toolbox and allow the user to apply the pre-trained models on new data. [ABSTRACT FROM AUTHOR]

In potato breeding, maturity class (MC) is a crucial selection criterion because this is a critical aspect of commercial potato production. Currently, the classification of potato genotypes into MCs is done visually, which is time- and labor-consuming. The objective of this research was to use vegetation indices (VIs) derived from unmanned aerial vehicle (UAV) imagery to remotely assign MCs to potato plants grown in trials, representing three different early stages within a multi-year breeding program. The relationships between VIs (GOSAVI – Green Optimized Soil Adjusted Vegetation Index, MCARI2 – Modified Chlorophyll Absorption Index-Improved, NDRE – Normalized Difference Red Edge, NDVI – Normalized Difference Vegetation Index, and OSAVI – Optimized Soil Adjusted Vegetation Index and WDVI – Weighted Difference Vegetation Index) and visual potato canopy status were determined. Further, this study aimed to identify factors that could improve the accuracy (decrease Mean Absolute Error – MAE) of potato MC estimation remotely. Results show that VIs derived from UAV imagery can be effectively used to remotely assign MCs to potato breeding lines, with higher accuracy for the potato B-clones (20 plants per plot) than the A-clones (6 plants per plot). Among the tested VIs, the NDRE allowed for potato MC evaluation with the lowest MAE. Applying NDRE for remote MC estimation using a validation dataset of potato B-clones (100 plants per plot), resulted in an MC estimate with a 0.81 MAE. However, the accuracy of potato MC estimation using UAV image-based methods should be improved by reducing the potato canopy's variability (increasing uniformity) within the plot. This could be achieved by minimizing 1) potato vines bending over the neighboring row, causing vine overlap between plots, and 2) plants damaged by tractor wheels during field operations. [ABSTRACT FROM AUTHOR]

Purpose: This paper proposes a low-cost and low-effort solution for determining the area of corn crops damaged by the wildlife facility utilising field images collected by an unmanned aerial vehicle (UAV). The proposed solution allows for the determination of the percentage of the damaged crops and their location. Methods: The method utilises image segmentation models based on deep convolutional neural networks (e.g., UNet family) and transformers (SegFormer) trained on over 300 hectares of diverse corn fields in western Poland. A range of neural network architectures was tested to select the most accurate final solution. Results: The tests show that despite using only easily accessible RGB data available from inexpensive, consumer-grade UAVs, the method achieves sufficient accuracy to be applied in practical solutions for agriculture-related tasks, as the IoU (Intersection over Union) metric for segmentation of healthy and damaged crop reaches 0.88. Conclusion: The proposed method allows for easy calculation of the total percentage and visualisation of the corn crop damages. The processing code and trained model are shared publicly. [ABSTRACT FROM AUTHOR]

Post‐release monitoring is critical for assessing translocation outcomes. Yet the quality of information gained from monitoring can vary greatly, and perceived monitoring costs often results in reduced monitoring effort. Selecting cost‐effective monitoring strategies that provide high quality data are therefore important for assessing translocation outcomes and making informed management decisions.To compare how monitoring strategy affects information gained, we monitored a toutouwai/North Island robin (Petroica longipes) reintroduction in Aotearoa New Zealand, based on monitoring objectives of determining survival, site fidelity and whether the extent of management was large enough to protect dispersing individuals. We compared how these objectives were met through four monitoring strategies: (1) comprehensive surveys with ground radio telemetry and playback; (2) aerial drone telemetry; (3) dedicated playback by trained surveyors and (4) opportunistic playback by predator control contractors. We undertook a viewshed analysis to determine search coverage of each strategy and compared detection rates, efficiency and cost.Comprehensive ground telemetry and playback, while costly, covered the largest area and provided the most accurate data on dispersal, survival and the translocation outcome. In comparison, opportunistic playback monitoring detected substantially fewer individuals, giving a false impression of low site fidelity and survival and a failed translocation. Although drone telemetry had considerable site‐specific limitations, which limited its effectiveness during our study, it was the most cost‐effective with a high detection rate and low search effort.Synthesis and applications: Our study shows the value of intensive monitoring in facilitating management decisions for wildlife translocations. Comprehensive telemetry and playback, while costly, were invaluable for gaining high quality information on the translocation outcome. Without suitable monitoring, reintroduction outcomes can be difficult to assess and potentially result in unnecessary, ineffective or overly expensive management actions. We recommend that monitoring intensity and methodology should reflect the site, species and level of uncertainty regarding the translocation outcome. Prioritising monitoring can help reduce long‐term costs, increase quality of information gained and allow for more informed management decisions that can improve subsequent translocation outcomes. [ABSTRACT FROM AUTHOR]

Mapping landslide cracks is essential in understanding landslide dynamics and evolution across space and time. Each landslide crack's depth, direction, and width are related to the stress and strain imposed on the landslide body. Moreover, their spatial distribution can indicate areas where the landslide can extend, mainly if located in the upper part of the main landslide scarp. Even though the cracks leave a distinct pattern on the landslide body when fresh or when there is a high contrast between the bare soil and surrounding vegetation, these patterns gradually diminish over time, making their detection difficult. The problem of landslide cracks mapping in various environmental conditions and having different ages was tackled in the current study using very high spatial resolution unmanned aerial vehicle (UAV) aerial imagery and derived products in conjunction with deep learning models. U-Net and DeepLab CNN models were applied using masked and non-masked training samples with different tile sizes. As the tile size decreases, the performance metrics, such as precision, recall, and F1-score, generally decrease. Overall, the lowest accuracy was approximately 0.79 for non-mask samples and tile size of 64 pixels, and reached over 0.93 for masked samples and tile size of 512 pixels. [ABSTRACT FROM AUTHOR]

The bridge models reconstructed from unmanned aerial vehicle (UAV) images via photogrammetry are often reported to have quality issues (e.g., high noise, insufficient resolution and precision loss) and thus restrict their application in bridge inspection. To address this problem, this paper proposes a novel Building Information Model (BIM)-based 3D path planning method for improving the quality of photogrammetric bridge models by optimising the UAV flight plan. This method firstly uses a simple BIM model as input and considers inspection and photogrammetry requirements to generate effective UAV viewpoints. Then it adjusts inaccessible UAV viewpoints that are in occupied space or cannot meet the flight safety rules. Finally, a feasible flight trajectory is generated through all valid viewpoints. To evaluate the performance of the proposed method, two prototypes were developed to automate the path planning and on-site image acquisition, respectively, and were tested on a real girder bridge. The results showed that, compared with the common UAV flight plan used in current practice, the proposed method could: (1) generate a more precise bridge model with fewer noise points and higher visual quality to support damage detection; and (2) significantly improve the efficiency of photogrammetric 3D bridge reconstruction with reduced human interventions in image collection and processing. [ABSTRACT FROM AUTHOR]

Summary: The present research work aims to investigate the reliability of the mixed free space optical (FSO)/radio frequency (RF) decode‐and‐forward (DF) relaying system where the satellite intends to communicate with the ground station through the unmanned aerial vehicles (UAV) as the relay node. Moreover, it has been considered that a second UAV interferes with the intended UAV. The operation of the UAVs has been represented considering the small scale fading, path loss, 3‐D location, and probability of maintaining line‐of‐sight (LoS) and non line of sight (NLoS) links, while FSO link undergoes Malaga distributed turbulence. Using the aforementioned model, the closed form expressions for the outage probability and bit error rate (BER) have been derived. The exact expressions have been extended to obtain the high signal‐to‐noise ratio (SNR) results for the outage probability and BER. The analytical expressions have been numerically evaluated and the results obtained through these expressions have been verified using the Monte Carlo simulations. More importantly, long short‐term memory (LSTM)‐based deep learning model has been trained for prediction of outage probability. The model is trained offline and then utilized to predict the values of the outage probability in online mode with mean square error (MSE) and root mean square error (RMSE) of MSE = −45.02 and RMSE = −18.34 dB. [ABSTRACT FROM AUTHOR]

Unmanned aerial vehicles (UAVs), which are available in our lives in many areas today, bring with them new expectations and needs along with developing technology. In order to meet these expectations and needs, main subjects such as reducing energy consumption, increasing thrust and endurance, must be taken into account in UAV designs. In this study, Backtracking search optimisation (BSO) algorithm-based adaptive neuro-fuzzy inference system (ANFIS) model is proposed for the first time to improve UAV thrust. For this purpose, first, different batteries and propellers were tested on the thrust measuring device and a data set was obtained. Propeller diameter and pitch, current, voltage and the electronic speed controller (ESC) signal were selected as input, and UAV thrust was selected as output. ANFIS was used to relate input and output parameters that do not have a direct relationship between them. In order to determine the ANFIS parameters at the optimum value, ANFIS was trained with the obtained data set by using BSO algorithm. Then, the objective function based on the optimum ANFIS structure was integrated into BSO algorithm, and the input values that gave the optimum thrust were calculated using BSO algorithm. Simulation results, in which parameters such as engine, battery and propeller affecting the thrust are taken into account equally, emphasise that the proposed method can be used effectively in improving the UAV thrust. This hybrid method, consisting of ANFIS and BSO algorithm, can reduce the cost and time loss in UAV designs and allows many possibilities to be tested. [ABSTRACT FROM AUTHOR]

The article focuses on the development of a fast and accurate method for forest fire detection using UAV imagery and Convolutional Neural Networks (CNNs). It highlights how the combination of high-resolution UAV images and deep learning techniques can automate the detection process, significantly reducing manual effort and errors while providing precise results.

Completing missions with autonomous decision-making unmanned aerial vehicles (UAV) is a development direction for future battlefields. UAV make decisions based on battlefield situation information collected by sensors and can quickly and accurately perform complex tasks such as path planning, cooperative reconnaissance, cooperative pursuit and attacks. Obtaining real-time situation information of enemy is the basis for realizing autonomous decision-making of the UAV. However, in practice, due to internal sensor failure or interference of enemy, the acquired situation information is prone to be missing, which affects the training and decision-making of autonomous UAV. In this paper, an adaptive missing situation data restoration method for UAV confrontation is proposed. The UAV confrontation situation data are acquired through JSBSim, an open-source UAV simulation platform. By fusing temporal convolutional network and long short-term memory sequences, we establish a deep regression method for missing data restoration and introduce an adaptive mechanism to reduce the training time of the restoration model in response to dynamic changes in the enemy’s strategy during UAV confrontation. In addition, we evaluate the reliability of the proposed method by comparing with different baseline models under different degrees of data missing conditions. The performance of our method is quantified by five metrics. The performance of our proposed method is better than the other benchmark algorithms. The experimental results show that the proposed method can solve the missing data restoration problem and provide reliable situation data while effectively reducing the training time of the restoration model. [ABSTRACT FROM AUTHOR]

In the event of a natural disaster such as a flood, cyclone, or tsunami, often the communication infrastructure gets damaged or becomes non-functional. In such scenarios, small cells covered by unmanned aerial vehicles (UAVs) are used to extend seamless cellular connectivity to the non-functional area (Mozaffari et al. in 2015 IEEE Global Communications Conference (GLOBECOM), IEEE, pp 1–6, 2015). This stand-in infrastructure is essential for sharing rescue-related information to the on-field rescue hubs and collecting real-time information from the devastating site to deliver to the outside world. Moreover, in such scenarios, it is possible that the Device-to-Device (D2D) nodes at the ground may have depleted battery power and hence are unable to transmit or receive information. In this paper, we propose a novel UAV-assisted multi-hop D2D communication model with energy harvesting for such a disaster scenario. In the proposed model, the best path from the UAV to the destination out of multiple D2D options is selected following a modified Dijkstra's algorithm. We develop analytical expressions for the end-to-end outage probability, ergodic capacity, and throughput in the multi-hop D2D links for making communication possible in the disaster zone. We evaluate the performance of the proposed scheme and observe significant improvements in terms of various network attributes. [ABSTRACT FROM AUTHOR]

Macroplastics are a global threat to the aquatic environment and will degrade into microplastics over time. Its presence in canal causes pollution and inhibits water flow, causing flooding in urban areas; therefore, it is essential to identify and monitor its presence. Addressing knowledge gaps is critical in determining solutions for mitigation purposes. In visual object detection studies, aerial mapping is developed with advanced technology, such as unmanned aerial vehicles (UAV). This research aims to conduct aerial mapping experiments to find the right formula or technical reference for detecting macroplastic waste objects floating on the surface of the canal, including flight altitude, exposure to sunlight, and the influence of season on object detection. Aerial mapping will be done in densely populated urban canals in Southeast Asia, Indonesia, and Makassar City. The aerial mapping survey method will be used, and then the data will be processed in the digitization process and object detection with GIS. The analysis kernel in GIS tools will be used to see the distribution density of macroplastics. The research results show that autoblock occurs at heights of 5m and 10m, but this autoblock can be minimized at a flight height of 15 m. Apart from that, height also affects flight duration. The lower flying height will result in better visual accuracy and better resolution. However, at a height of 15m, macroplastic objects were still detected on a moderate scale. This research successfully distinguished various plastic waste materials, the most found being the soft polyolefin category in plastic bags. Monitoring results detected 321 items of macroplastics in the dry season and 1,163 in the rainy season, or a threefold increase with conditions spread thinly in the dry season. In the rainy season, they gather densely on one side of the canal. [ABSTRACT FROM AUTHOR]

The increasing use of drone technology to produce high-resolution digital imagery and elevation models has been associated with a growing interest in developing quantitative morphometric analysis (QMA). QMA analysis is an invaluable part of creating detailed topographic models in landslide scars that are still highly unstable and prone to erosion. This paper presents the results of a research that aims to create a topographic model in a landslide scarred area where the slope configuration is still varied. The study area was located in the landscape of the Cretaceous-Tertiary volcanic transition where many landslides have occurred. Three landslides were selected on the basis of different soil material characteristics that affect the topographic condition of the landslide scar. Aerial photography was recorded using a UAV with a flying height of 80 m, with an orthomosaic resolution of 1 cm. In detail, three morphometric variables (slope, plan curvature, topographic position index) were selected and calculated with the output evaluated based on visual-spatial interpretation. The results showed that morphometric variables performed well in modeling land surface topography. Steep slopes and surfaces with convex curvature are abundant at the ledges and landslide heads that allow water runoff to disperse as the initiation of gully erosion. The multidimensional gully erosion network is concentrated at relatively low elevations and surfaces with concave curvature. The undulating micro-relief of the land surface as a result of the process of material disposition builds up on each other to a gentle slope. Finally, the topographic model of the landslide surface can be used as a base material in implementation of both physical and vegetative land conservation strategies. [ABSTRACT FROM AUTHOR]

Non-invasive crop analysis through image-based methods holds great promise for applications in plant research, yet accurate and robust trait inference from images remains a critical challenge. Our study investigates the potential of AI model ensembling and hybridization approaches to infer sorghum crop traits from RGB images generated via unmanned aerial vehicle (UAV). In our study, we cultivated 21 sorghum cultivars in two independent seasons (2021 and 2022) with a gradient of fertilizer and water inputs. We collected 470 ground-truth N measurements and captured corresponding RGB images with a drone-mounted camera. We computed five RGB vegetation indices, employed several ML models such as MLR, MLP, and various CNN architectures (season 2021), and compared their prediction accuracy for N-inference on the independent test set (season 2022). We assessed strategies that leveraged both deep and handcrafted features, namely hybridized and ensembled AI architectures. Our approach considered two different datasets collected during the two seasons (2021 and 2022), with the training set from the first season only. This allowed for testing of the models' robustness, particularly their sensitivity to concept drifts, in the independent season (2022), which is fundamental for practical agriculture applications. Our findings underscore the superiority of hybrid and ensembled AI algorithms in these experiments. The MLP + CNN-VGG16 combination achieved the best accuracy (R2 = 0.733, MAE = 0.264 N% on an independent dataset). This study emphasized that carefully crafted AI-based models applied to RGB images can achieve robust trait prediction with accuracies comparable to the similar phenotyping tasks using more complex (multi- and hyper-spectral) sensors presented in the current literature. [ABSTRACT FROM AUTHOR]

This paper discusses modelling of a multibody system consisting of airship, gondola, and a slung payload. Lighter-than-air vehicles undergo inertial forces that are often neglected in heavier-than-air vehicles. These inertial forces are modelled using added mass and added inertia. The dynamics of the multibody system were first modelled using the Udwadia–Kalaba method. Three constraints were derived and enforced. The resulting equation of motion was used to identify the added mass, added inertia, and inertia of the airship through system identification procedure. The proposed system identification method utilizes semidefinite programming with equality and inequality constraints to find any unknown parameters in the mass matrix of the multibody system. Three experiments were carried out to perform the system identification and validate the dynamic model. The identified mass matrix was used to reconstruct the trajectories of the experiments. Using the experimentally obtained mass matrix demonstrated 35 % lower error when compared with simulated trajectories using approximated mass matrices. [ABSTRACT FROM AUTHOR]

The article presents a method of planning the flight trajectory of a swarm of drones using a modified RRT (Rapidly-exploring Random Tree) algorithm. The version of the RRT algorithm presented in the article is called Multiplatform Spacetime RRT*. The proposed modifications make it possible to determine the flight trajectory of UAVs taking into account time constraints related to the area occupied by each platform. Additionally, the proposed algorithm ensures the avoidance of potential collisions of platforms in the air by using a collision avoidance algorithm used in practice based on geometric methods. Two designed and tested modifications of RRT were presented, based on the basic RRT* and Informed RRT* algorithms. The algorithm used in both tested versions guarantees the determination of the optimal flight path for unmanned platforms in a finite, small number of steps, which solely depends on the number of UAVs involved. This algorithm takes into account the dynamic model of the fixed-wing UAV. The simulation results presented by planning the flight trajectory of a swarm, consisting of three and four UAVs using the Multiplatform Spacetime RRT* algorithm, are significantly better than the algorithms that were compared to achieve these results. [ABSTRACT FROM AUTHOR]

Context: The invasion of annual grasses in western U.S. rangelands promotes high litter accumulation throughout the landscape that perpetuates a grass-fire cycle threatening biodiversity. Objectives: To provide novel evidence on the potential of fine spatial and structural resolution remote sensing data derived from Unmanned Aerial Vehicles (UAVs) to separately estimate the biomass of vegetation and litter fractions in sagebrush ecosystems. Methods: We calculated several plot-level metrics with ecological relevance and representative of the biomass fraction distribution by strata from UAV Light Detection and Ranging (LiDAR) and Structure-from-Motion (SfM) datasets and regressed those predictors against vegetation, litter, and total biomass fractions harvested in the field. We also tested a hybrid approach in which we used digital terrain models (DTMs) computed from UAV LiDAR data to height-normalize SfM-derived point clouds (UAV SfM-LiDAR). Results: The metrics derived from UAV LiDAR data had the highest predictive ability in terms of total (R2 = 0.74) and litter (R2 = 0.59) biomass, while those from the UAV SfM-LiDAR provided the highest predictive performance for vegetation biomass (R2 = 0.77 versus R2 = 0.72 for UAV LiDAR). In turn, SfM and SfM-LiDAR point clouds indicated a pronounced decrease in the estimation performance of litter and total biomass. Conclusions: Our results demonstrate that high-density UAV LiDAR datasets are essential for consistently estimating all biomass fractions through more accurate characterization of (i) the vertical structure of the plant community beneath top-of-canopy surface and (ii) the terrain microtopography through thick and dense litter layers than achieved with SfM-derived products. [ABSTRACT FROM AUTHOR]

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Przegląd Elektrotechniczny is the property of Przeglad Elektrotechniczny and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Journal of Agricultural Science & Technology (1008-0864) is the property of Journal of Agricultural Science & Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Pine wilt disease (PWD) is a severe forest disease caused by the invasion of pine wood nematode (Bursaphelenchus xylophilus), which has caused significant damage to China's forestry resources due to its short disease cycle and strong infectious ability. Benefiting from the development of unmanned aerial vehicle (UAV)-based remote sensing technology, the use of UAV images for the detection of PWD-infected trees has become one of the mainstream methods. However, current UAV-based detection studies mostly focus on multispectral and hyperspectral images, and few studies have focused on using red–green–blue (RGB) images for detection. This study used UAV-based RGB images to extract feature information using different color space models and then utilized semantic segmentation techniques in deep learning to detect individual PWD-infected trees. The results showed that: (1) The U-Net model realized the optimal image segmentation and achieved the highest classification accuracy with F1-score, recall, and Intersection over Union (IoU) of 0.9586, 0.9553, and 0.9221, followed by the DeepLabv3+ model and the feature pyramid networks (FPN) model. (2) The RGBHSV feature scheme outperformed both the RGB feature scheme and the hue saturation value (HSV) feature scheme, which were unrelated to the choice of the semantic segmentation techniques. (3) The semantic segmentation techniques in deep-learning models achieved superior model performance compared with traditional machine-learning methods, with the U-Net model obtaining 4.81% higher classification accuracy compared with the random forest model. (4) Compared to traditional semantic segmentation models, the newly proposed segment anything model (SAM) performed poorly in identifying pine wood nematode disease. Its success rate is 0.1533 lower than that of the U-Net model when using the RGB feature scheme and 0.2373 lower when using the HSV feature scheme. The results showed that the U-Net model using the RGBHSV feature scheme performed best in detecting individual PWD-infected trees, indicating that the proposed method using semantic segmentation technique and UAV-based RGB images to detect individual PWD-infected trees is feasible. The proposed method not only provides a cost-effective solution for timely monitoring forest health but also provides a precise means to conduct remote sensing image classification tasks. [ABSTRACT FROM AUTHOR]

Context: Coastal beach environments provide habitats for marine megafauna, including turtles, rays, marine mammals and sharks. However, accessing these variable energy zones has been difficult for researchers by using traditional methods. Aims: This study used drone-based aerial surveys to assess spatio-temporal variation of marine megafauna across south-eastern Queensland, Australia. Methods: Drones were operated at five south-eastern Queensland beaches. Megafauna sightings and key variables including location, month and turbidity were analysed to assess variation across locations. Key results: Overall, 3815 individual megafauna were detected from 3273 flights. There were significant differences in the composition of megafauna assemblages throughout the year and among beaches, with megafaunal sightings in >80% of flights conducted off North Stradbroke Island. Conclusions: Strong temporal presence was found that is congruent with other studies examining seasonality. This supports the use of drones to provide ecological data for many hard-to-study megafauna species and help inform long-term sustainable management of coastal ecosystems. Implications: Results indicated that environmental conditions can influence the probability of sighting marine megafauna during aerial surveys. Studying beach environments by using boat-based methods has been difficult. Drone surveys were used to assess abundance patterns of large marine animals in south-eastern Queensland, Australia. Significant differences in the composition of animal groups were found throughout the year and among locations, with large marine animal sightings in >80% of flights conducted off North Stradbroke Island. This supports the use of drones to provide scientific data for hard-to-study marine species and inform management of coastal ecosystems. [ABSTRACT FROM AUTHOR]

The current paper discusses the application of Sliding Mode Control (SMC) to a quadcopter. The controller is designed based on the system's nonlinear model. An adaptive sliding mode controller is developed specifically for the quadcopter's attitude subsystem, aiming to mitigate the undesirable vibration phenomena typically associated with conventional sliding mode controllers while ensuring robust trajectory tracking for the quadcopter's attitude. The stability of the proposed controller was verified using the Lyapunov stability theorem. The quadcopter Unmanned Aerial Vehicle (UAV) model and the performance of the proposed controller were simulated and validated in MATLAB/SIMULINK environment. The results demonstrate that the proposed controller effectively positions the quadcopter with minimal error, maintaining the UAVs flight along the prescribed trajectory. Additionally, it performs well in trajectory tracking under collision noise and vibration reduction conditions. [ABSTRACT FROM AUTHOR]

Copyright of Xinjiang Agricultural Sciences is the property of Xinjiang Agricultural Sciences Editorial Department and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The geometric correction of thermal infrared (TIR) orthophotos generated by unmanned aerial vehicles (UAVs) presents significant challenges due to low resolution and the difficulty of identifying ground control points (GCPs). This study addresses the limitations of real-time kinematic (RTK) UAV data acquisition, such as network instability and the inability to detect GCPs in TIR images, by proposing a method that utilizes RGB orthophotos as a reference for geometric correction. The accelerated-KAZE (AKAZE) method was applied to extract feature points between RGB and TIR orthophotos, integrating binary descriptors and absolute coordinate-based matching techniques. Geometric correction results demonstrated a significant improvement in regions with stable and changing environmental conditions. Invariant regions exhibited an accuracy of 0.7~2 px (0.01~0.04), while areas with temporal and spatial changes saw corrections within 5~7 px (0.10~0.14 m). This method reduces reliance on GCP measurements and provides an effective supplementary technique for cases where GCP detection is limited or unavailable. Additionally, this approach enhances time and economic efficiency, offering a reliable alternative for precise orthophoto generation across various sensor data. [ABSTRACT FROM AUTHOR]

Research on panicle detection is one of the most important aspects of paddy phenotypic analysis. A phenotyping method that uses unmanned aerial vehicles can be an excellent alternative to field-based methods. Nevertheless, it entails many other challenges, including different illuminations, panicle sizes, shape distortions, partial occlusions, and complex backgrounds. Object detection algorithms are directly affected by these factors. This work proposes a model for detecting panicles called Border Sensitive Knowledge Distillation (BSKD). It is designed to prioritize the preservation of knowledge in border areas through the use of feature distillation. Our feature-based knowledge distillation method allows us to compress the model without sacrificing its effectiveness. An imitation mask is used to distinguish panicle-related foreground features from irrelevant background features. A significant improvement in Unmanned Aerial Vehicle (UAV) images is achieved when students imitate the teacher’s features. On the UAV rice imagery dataset, the proposed BSKD model shows superior performance with 76.3% mAP, 88.3% precision, 90.1% recall and 92.6% F1 score. [ABSTRACT FROM AUTHOR]

Palsas and peat plateaus in subarctic peatlands are some of the southernmost lowland permafrost landforms in the Northern Hemisphere. Peatland permafrost along the Labrador Sea coastline in northeastern Canada is substantially understudied despite the importance of these landforms for wildlife, carbon stores, and Indigenous land users. In this study, we derived geomorphological and resiliency indices for peatland permafrost landforms at 20 wetland complexes, spanning a latitudinal gradient from Blanc‐Sablon, QC (51.4° N), to Nain, NL (56.5° N). Orthomosaics and three‐dimensional point clouds were created for each site using UAV‐based surveys and structure‐from‐motion photogrammetry. Analyses revealed that landforms in the region are characterized by short heights (maximum height: 3.65 m, average height: 0.49 m) with lichen and dwarf shrub cover, making them more similar to features in northern Europe than western Canada. Palsas and peat plateaus ranged in size from 49 to 14,233 m2, with a median feature size of 259 m2. Peatland permafrost in the region exhibits high levels of fragmentation, with most study sites (90%) exhibiting low or very low thaw resiliency. The results from this study indicate a high vulnerability to thaw and degradational processes, with potential negative consequences for species of cultural value to Labrador Inuit, Kallunângajuit, and Innu. [ABSTRACT FROM AUTHOR]

In modern warfare, the use of UAVs for reconnaissance, search and rescue missions is very common, and it is essential to plan the flight path of UAVs. However, in the face of complex battlefield environment, the existing flight path planning algorithms have the problems of long time consumption and unstable path. Therefore, this paper studies the UAV flight path planning optimization in complex battlefield environment. First, we construct the battlefield environment model. Then, by analyzing the UAV flight constraints existing in battlefield environment, the objective function is obtained. And the problem of UAV flight path planning optimization is transformed into a nonlinear combinatorial optimization problem. On this basis, an Adaptive Adjustment Flight Path Planning algorithm (AA-FPP) is proposed. The AA-FPP algorithm adaptively adjusts the absorption coefficient of fireflies by using chaotic strategy. It adjusts the control position updating formula by using time-varying inertia weight to enhance its global searching ability. Then, random factors based on Boltzmann selection strategy are introduced to perturb the iterative solutions in AA-FPP. It expands the search space of the path and enhances the convergence efficiency. Finally, simulation results show that the AA-FPP algorithm can successfully plan a flight path that reduces static/dynamic threat intensity. And it has greater advantages in path stability and planning time consumption. [ABSTRACT FROM AUTHOR]

Uncrewed Aerial Vehicles (UAVs; commonly known as drones) have been gaining interest as a potential transport mode for logistics (i.e., payload delivery), bringing suggested benefits such as reduced transit times and improved access in hard-to-reach locations. However, drones have yet to become widely established in routine logistics systems, with a postulated reason being that the higher costs associated with operating drones relative to traditional surface transport modes (e.g., vans, bicycle couriers) present a barrier to uptake. Based on case studies of two real-world logistics networks transporting patient pathology samples in a healthcare setting in the United Kingdom, this study investigated the effects of the relative costs of drones on mode choice in integrated, multi-modal logistics systems. Results suggested that drones could be a financially viable option if their costs reduced below ∼19% of current values, although such a reduction may not be feasible, even in a future involving increased drone automation. Drones reduced sample transit times by up to ∼70% compared to vans but benefits to the wider healthcare system were negligible because level of service requirements for transit times could be achieved by all modes.

As the world continues to urbanize at an unprecedented rate, the energy demand in cities is rising. Buildings account for over 75% of all the energy consumed in cities and are responsible for over two-thirds of the emissions. Assessment of energy demand in buildings is a highly integrative endeavour, bringing together the interdisciplinary fields of energy and urban studies, along with a host of technical domains namely, geography, engineering, economics, sociology, and planning. In the last decade, several urban building energy modelling tools (UBEMs) have been developed for estimation as well as prediction of energy demand in cities. These models are useful in policymaking as they can evaluate future urban energy scenarios. However, data acquisition for generating the input database for UBEM has been a major challenge. In this review, a comprehensive assessment of the potential of remote sensing and GIS techniques for UBEM has been presented. Firstly, the most common input variables of UBEM have been identified by reviewing recent publications on UBEM and then studies related to the acquisition of data corresponding to these variables have been explored. More than 140 research papers and review articles relevant to remote sensing and GIS applications for building level data extraction in urban areas and UBEM applications have been investigated for this purpose. After going through level of details required for each of the input components of UBEM and studying the possibility of acquiring some of those data using remote sensing, it has been inferred that satellite remote sensing and Unmanned Aerial Vehicles (UAVs) have a strong potential in enhancing the input data space for UBEM but their applicability has been limited. Further, the challenges of the usage of these technologies and the possible solutions have also been presented in this study. It is recommended to utilise the existing methodologies of extracting information from remote sensing and GIS for UBEM, along with newer techniques such as machine learning and artificial intelligence.

Abstract The allocation of communication resources is extremely important for the device-to-device (D2D) communication network with unmanned aerial vehicle (UAV) relays, but this field currently mainly focuses on the communication between UAVs and ground devices. While neglecting the comprehensive optimization of air-to-ground communication and internal ground communication. This paper studies the optimization problem of communication for ground D2D devices based on UAV relays in urban environment. We study the comprehensive optimization considering the air-to-ground communication and the mobile ground communication devices. We construct the D2D communication neighbor table, and the terrestrial D2D devices will sort and select the best D2D link for pairing based on the Euclidean distance and gain set around them. We construct a communication model for this scenario and propose an improved Dual-Actor Network Softmax Twin Delayed Deep Deterministic Policy Gradient algorithm based on deep reinforcement learning. We perform transmit power optimization to address the interference caused by D2D communication using frequency division multiplexing technology. By designing the reward function and training mechanism, we can achieve good power control and reduce communication interference. The results of the simulation analysis performed in this scenario show an effective reduction in communication latency and an increase in D2I (device-to-infrastructure) channel capacity, D2D connectivity success, and energy efficiency. The feasibility and effectiveness of the algorithm is verified by comparing the baseline algorithm.

Abstract A mask identification and social distance monitoring system using Unmanned Aerial Vehicles (UAV) in the outdoors has been proposed for a health establishment. The above approach performed surveillance of the surrounding area using cameras installed in UAVs and internet of things technologies, and the captured images seem useful for tracking the entire environment. However, innate images from unmanned aerial vehicles show an adaptable visual effect in an uncontrolled environment, making face-mask detection and recognition harder. The UAV picture first had to be converted to grayscale, then its contrast was amplified. Image contrast was improved using Optimum Wavelet-Based Masking and the Enhanced Cuckoo Methodology (ECM). According to the contrast-enhanced image, Gabor-Transform (GT) and Stroke Width Transform (SWT) methods are used to derive attributes that help categorise mask-wearers and non-mask-wearers. Using the retrieved attributes, a Weighted Naive Bayes Classification (WNBC) detected masks in the images. Additionally, a deep neural network-based, the faster Region-Based Convolutional Neural Networks (R-CNN) algorithm combined with Adaptive Galactic Swarm Optimization (AGSO) is being used to identify appropriate and incorrect face mask wear in images, as well as to monitor social distancing among individuals in crowded areas. When the system recognises unmasked individuals, it sends their information to the doctor and the nearby police station. One unmanned aerial vehicle’s automated system alert people via speakers, ensuring social spacing. The problem involves a large percentage of appropriate and incorrect face mask wear using data from GitHub and Kaggle, including a training repository of 16,000 images and a testing data set of 12,751 images. To enhance the performance of the model’s learning, the methodology of 10-fold cross-validation will be used. Precision, recall, F1-score, and speed are then measured to determine the efficacy of the suggested approach.

Drained peatlands need to be rewetted to reduce carbon emissions. To realise this, sustainable land-use alternatives must be implemented after rewetting. Uncrewed Aerial Systems (UAS) can aid in monitoring crop growth and spatial heterogeneity in vegetation patterns, and hence contribute to improved management. We monitored Typha latifolia (‘Typha’) biomass for an 8.5 ha rewetted paludiculture site in north-east Germany using structural (digital surface model, DSM) and multispectral data (5 spectral bands and normalized difference vegetation index, NDVI) obtained by drone surveys in July, August and September 2021. We used in-situ harvests of Typha from 1 m2 square plots as training data. Biomass for validation plots was predicted from field measurements for the respective observation dates. The DSM’s and NDVI’s spatial resolution (2.76 cm) were resampled to 1 m and original values aggregated into different spatial metrics (i.e., percentiles of pixel height and NDVI values). Different regression models were separately tested for the different August DSM and NDVI metrics as explanatory variables from August data. A normalised Typha fraction cover mask from the multispectral data was used to exclude non-target species. To test the model for different phenological stages, we then applied the best performing model of August to July and September. The models were compared to non-destructive biomass predictions from linear relationships between field measurements. The combination of DSM or NDVI metrics with the Typha mask captured the heterogeneity of Typha biomass well (R2 = 0.65–0.71). Biomass overprediction for present non-target species was successfully excluded. DSM models outperformed NDVI models for dense Typha stands due to saturation of the NDVI at 500 g m-2 biomass. We were able to show biomass accumulation from July to August of up to 200 g m-2. The September model had the lowest performance (R2 = 0.6), due to a weakened height-biomass relation. Further, our model underestimated flower biomass. As single UAS surveys offer both structural and spectral information, UAS data will contribute to precise biomass and vegetation monitoring at high spatial resolution in upcoming rewetting efforts.

Unmanned aerial vehicles (UAVs) are primarily used in the field of cultural heritage for mapping the exteriors of larger objects and documenting the roofs and façades of tall structures that cannot be efficiently or feasibly measured using conventional terrestrial technologies and methods. However, due to the considerable diversity of cultural heritage, there are practical demands for the measurement of complex and inaccessible objects in interior spaces. This article focuses on the use of two different off-the-shelf UAVs for partial photogrammetric reconstruction of the attic of a mining house, which was only visible through a window in the gable wall. Data from both UAVs were compared with each other and with terrestrial laser scanning. Despite the lower quality of the results from the DJI Mini 4 Pro compared to the DJI Mavic 3 Enterprise, the results from both UAVs would still be suitable for documenting the interior attic spaces. However, a detailed analysis of the photogrammetric data indicates that, when selecting a UAV for this purpose, it is necessary to consider the limitations of the camera system, which may lead to a reduction in the geometric accuracy and completeness of the point clouds.

【Objective】 The leaf area index (LAI) is an important trait of plant canopies but challenging to measure accurately at large scales. We studied the feasibility of using multispectral imaging and machine learning to estimate the LAI of spring wheat. 【Method】 The experiment was conducted in a spring wheat field on the Tumochuan Plain in the Yellow River Basin, Inner Mongolia. Images of the spring wheat at the jointing, booting, and grain-filling stages were acquired using a multispectral camera mounted on a DJI P4M UAV. Selected vegetation indices were subjected to principal component analysis (PCA), and the resulting components were used to estimate LAI. We compared six models: multiple linear regression (MLR), decision tree regression (DTR), backpropagation neural network regression (BPNN), gradient boosting decision tree regression (GBDT), support vector machine regression (SVR), and random forest regression (RFR). LAI was calculated separately for each growth stage using different vegetation indices. 【Result】 LAI was significantly correlated with the normalized difference vegetation index (NDVI), modified simple ratio (MSR), ratio vegetation index (RVI), difference vegetation index (DVI), soil-adjusted vegetation index (SAVI), and normalized difference red edge index (NDRE). It showed a weak correlation with the renormalized difference vegetation index (RDVI) during the heading and grain-filling stages, with their correlation coefficients being 0.23 and 0.21, respectively. The BPNN model was most accurate during the jointing stage, with R2, RMSE, and MAE being 0.822, 0.305, and 0.257, respectively. In contrast, the RFR model performed best during the heading, grain-filling, and entire growth periods, with R2 being 0.613, 0.811 and 0.834, RMSE being 0.189, 0.150 and 0.174, and MAE being 0.126, 0.121 and 0.133, respectively. Additionally, the RFR model constructed using data from all three stages was more accurate than models derived from data at individual growth stages. 【Conclusion】 Multispectral data acquired via UAV, combined with machine learning algorithms, can accurately estimate the LAI of spring wheat at various growth stages. Models constructed using data from multiple growth stages are more accurate than those based on a single stage. The models are most accurate for the booting stage and least for the heading stage. Overall, the RFR model provided the most accurate LAI estimates across the three growth stages.

Abstract Blockchain is a decentralized electronic account book that is secure, open, and impervious to manipulation, and records transactions across numerous computers. Internet of Things (IoT) deployed sensors are limited in bandwidth, processing power, and battery life. Energy efficiency is a vital factor and a decisive matter for energy-bound IoT-based networks. This paper provides an idea of the working principle of blockchain technology in IoT-enabled energy systems. It investigates the fundamentals of blockchain technology, clarifying its decentralized nature, cryptographic mechanisms, and consensus algorithms that ensure data immutability and transparency. It explores how blockchain technology can be integrated with various IoT fields for energy efficiency. The paper investigates the blockchain-based energy consumption approaches in IoT qualitatively, and quantitatively with future research directions and challenges. It is seen that the hybrid blockchain model is suitable for energy efficiency in IoT.

Abstract The rapid development of UAV communication technology makes it have application potential in wireless systems. However, for the optimization problem of UAV base station providing communication for ground personnel in the disaster area under special natural disasters, traditional deep reinforcement learning is used. Algorithms cannot solve such problems well. This article proposes the AD3QN algorithm combined with the attention mechanism, which can communicate with disaster victims on the ground better and more quickly, providing better information collection for rescue missions and completing the task of communication optimization. In the mission simulation environment, in order to make the environment more realistic, we innovatively designed the ground user distribution model, air-to-ground communication model, and electromagnetic interference model. Finally, the effect of the proposed algorithm is evaluated by comparing different deep reinforcement learning algorithms. The results show that the algorithm we proposed can provide communication services faster and has higher practicability in terms of algorithm.

Abstract Land Use/Cover Change (LUCC) plays a crucial role in sustainable land management and regional planning. However, contemporary feature extraction approaches often prove inefficient at capturing critical data features, thereby complicating land cover categorization. In this research, we introduce a new feature extraction algorithm alongside a Segmented and Stratified Principal Component Analysis (SS-PCA) dimensionality reduction method based on correlation grouping. These methods are applied to UAV LiDAR and UAV HSI data collected from land use types (e.g., residential areas, agricultural lands) and specific species (e.g., tree species) in urban, agricultural, and natural environments to reflect the diversity of the study area and to demonstrate the ability of our methods to be applied in different classification scenarios. We utilize LiDAR and HSI data to extract 157 features, including intensity, height, Normalized Digital Surface Model (nDSM), spectral, texture, and index features, to identify the optimal feature subset. Subsequently, the best feature subset is inputted into a random forest classifier to classify the features. Our findings demonstrate that the SS-PCA method successfully enhances downscaled feature bands, reduces hyperspectral data noise, and improves classification accuracy (Overall Accuracy = 91.17%). Additionally, the CFW method effectively screens appropriate features, thereby increasing classification accuracy for LiDAR (Overall Accuracy = 78.10%), HIS (Overall Accuracy = 89.87%), and LiDAR + HIS (Overall Accuracy = 97.17%) data across various areas. Moreover, the integration of LiDAR and HSI data holds promise for significantly improving ground fine classification accuracy while mitigating issues such as the ‘salt and pepper noise’. Furthermore, among individual features, the LiDAR intensity feature emerges as critical for enhancing classification accuracy, while among single-class features, the HSI feature proves most influential in improving classification accuracy.

In this work, a group of unmanned aerial vehicles (UAVs) is aimed to function as relay nodes that provide a multi-hop wireless link between ground users. The communication route between the sender and receiver is established with awareness of the link path loss between the cooperating UAV relay nodes. Then, Dijkstra’s shortest path algorithm is used to find the optimal route that has the minimum total path loss among all other possible routes. MATLAB simulation results show a high-performance gain when using Dijkstra’s algorithm for establishing the communication route compared with the baseline approach in which the route is formed based on the shortest-distance path through the UAV relay nodes. It is worth mentioning that both shadow fading and nonfading radio environments were taken into account. Moreover, the performance is further improved when the standard deviation of shadow fading increases. The work also investigates the probability of selecting each of the UAV nodes in forming the multi-hop link considering the total simulated realizations. Here, it is shown that more variability is observed in the indexes of the traversed UAV nodes when routeing through a random fading environment compared to a nonfading environment.

This study examines a cooperative non-orthogonal multiple access (NOMA) network utilizing an energy-constrained unmanned aerial vehicle (UAV) relay (UAVR) to expand coverage and improve network throughput. In order to provide energy to the UAVR, we consider the hybrid simultaneous wireless information and power transmission (SWIPT) method, which allows the UAVR to harvest energy from the source (i.e., sink node) signal. Herein, a hybrid time switching (TS)-based and power splitting (PS)-based relaying scheme is applied to improve the UAVR’s energy harvesting (EH) efficiency and the system performance. Given this context, we derive the closed-form expression of the outage probability (OP) for the sensors to evaluate the network performance. Based on the achieved analytical results, we apply the bat algorithm optimization (BAO) method to determine the optimal working point (as a fraction of received power and power allocation coefficients, and the 3-D positions of the UAVR) for the system such that the OP is minimized. The numerical analysis indicates that BAO is effective in both exploring and exploiting solutions, making it a suitable choice for similar non-convex optimization problems in continuous search spaces for cooperative NOMA IoT networks.

Devices that are employed in applications related to the Internet of Things (IoT) are constrained by limited energy resources. Consequently, ensuring a continuous supply of energy while also maintaining uninterrupted connectivity within IoT units (IoTUs) is of great importance. In this particular context, we present a scheme that facilitates both, the transfer of wireless power and the transmission of information for IoTUs along with the capability of harvesting solar energy. This scheme is further supported by the utilization of unmanned aerial vehicles (UAV) and the deployment of reconfigurable intelligent surfaces (RIS) for communication purposes. To be more precise, initially, IoTUs obtain energy from the UAV through the process of wireless power transmission (WPT). Subsequently, in the second stage, the UAV retrieves data from the IoTUs using transmitting information. In order to simplify the complexity of the communication issue, we assume that a solar-powered UAV remains stationary at a predetermined altitude. Our objective is to maximize the energy efficiency (EE) of the entire network by coordinating the scheduling of IoTU energy harvesting (EH) and UAV trajectory optimization. We suggest a multi-agent federated reinforcement learning (MFRL) algorithm that maximizes EE through parameter optimization in order to achieve this goal. By utilizing the collective experiences of several agents and reducing energy usage, this algorithm also improves the overall performance of the system. The proposed technique achieves 96.3% and 97.5% accuracy in communication rounds and RIS elements, with a 9−33% increase in EE compared to the best-performing benchmark scheme. The suggested approach outperforms the benchmark algorithms in terms of EE, trajectory optimization, and learning accuracy, according to simulation findings.

Abstract The operation mode of UAV equipped with LiDAR can significantly reduce the efficiency and cost of highway terrain acquisition. At present, due to the complicated terrain and weak signal of mountain road, it is difficult to collect terrain data. In view of the above problems, this paper puts forward the technology of BDS assisted terrain acquisition and processing on complex mountain roads. This paper describes in detail the new signal unfuzzy acquisition and tracking algorithm and multi-antenna positioning and attitude determination technology and equipment of BDS-3 mountain highway, through millimeter wave Lidar altimetry technology, imitation ground flight, air-ground multi-source data combined with air three-direction method and other technologies. To realize efficient 3D collection and 3D reconstruction of massive original images of terrain of complex mountain traffic lines in the range of 100 km. Finally, a case study of 100 km mountain road is taken to verify the effectiveness of the method.

Abstract. The development and application of automated technologies in the agricultural sector have significantly transformed the management and monitoring of farm animals. This paper presents a comprehensive analysis of existing technical systems for measuring the physiological parameters of farm animals, particularly focusing on cows. The primary goal of these technologies is to enhance the accuracy and efficiency of health monitoring and to optimize livestock management through continuous data collection and analysis. Key technological solutions include the use of wireless data transmission systems, radio monitoring devices, and unmanned aerial vehicles (UAVs). These technologies enable real-time tracking of animal health parameters such as body temperature, heart rate, movement patterns, and other critical indicators. The use of various sensors, including ear, neck, and intra-body sensors, plays a crucial role in collecting primary physiological data. These sensors are integrated into a broader system that includes radio modules with built-in controllers for converting raw data into digital formats, enabling wireless transmission to central systems for processing and storage. The radio monitoring of animals allows for the continuous tracking of physiological and behavioral parameters across large areas, providing crucial insights into their well-being. It is particularly beneficial in detecting early signs of disease or health issues, ensuring timely intervention, and preventing potential economic losses for farmers. The integration of unmanned aerial vehicles (UAVs) further expands the capabilities of monitoring systems, enabling the visual assessment of animals' behavior and movement across vast pastures. Equipped with video cameras and data relays, UAVs collect video footage that is transmitted to a central processing unit for real-time analysis and storage. The combined use of radio monitoring and UAV technology forms a comprehensive framework for livestock management, allowing farmers to make data-driven decisions, optimize resource allocation, and improve overall herd health and productivity. By leveraging these automated systems, it becomes possible to reduce the dependency on manual labor, minimize human error, and ensure that animals are closely monitored without the need for constant physical presence. This represents a significant advancement in modern animal farming, aligning with global trends toward smart farming and precision agriculture. In conclusion, the integration of automated physiological monitoring technologies in farm animal management presents an innovative solution for enhancing animal welfare, improving farm efficiency, and reducing operational costs. These systems hold considerable potential for further advancements in the field of animal health monitoring, ultimately contributing to more sustainable and productive farming practices.

Objective. The goal is to develop a web service for detecting aerial objects that detects a flying object, highlights it in an image and classifies it as a threat, since modern aerial object detection systems do not always cope with the task of detecting unmanned aerial vehicles due to their small size, low flight altitude and the use of materials that are barely noticeable to radar stations. Unmanned aerial vehicles that operate without operator control make it difficult to detect by radio signals. To detect UAVs, it is proposed to use a system based on optical scanning of the sky around protected objects. The system should be capable of autonomous operation and include aerial object detectors created on the basis of computer vision and artificial intelligence technologies. Method. The research and development of the airspace monitoring web service are based on the methods of system analysis, synthesis, and deduction. Result. The visual part of the web interface has been designed and developed; a dataset has been formed from open sources for the correct detection of flying objects; a neural network detector has been developed for classifying flying objects that pose a danger; a software module has been developed that allows for the automatic detection of identification flags of dangerous air objects with subsequent provision of reports in txt files in yolo format (coordinates are normalized). Conclusion. Separation of the visual part of the service will allow for distributed deployment of the server part, increasing flexibility and scalability. Development of the administrator control panel will allow for effective control of the service, management of settings and users. As a result, the web service will be able to: monitor the sky around protected objects, automatically detecting and classifying air objects and identifying air objects by threat level, providing information for taking necessary measures.

As climate change intensifies, extreme weather events like floods are occurring with increasing frequency. While data-driven deep learning methods are effective for extracting flood disaster information, their efficiency is constrained by the scarcity of postdisaster samples, the high cost of annotations, and the models’ strong dependence on both the quantity and quality of data. This study introduces an interactive semantic segmentation model based on multisource UAV flood images, incorporating four types of prompts. By embedding expert knowledge into the prompt design, the model reduces annotation costs and enhances generalization capabilities. First, a prompt encoder is developed to map different types of prompt information into a three-channel space using convolutional techniques, thereby reducing sample labeling costs. Moreover, an image encoder that integrates Mamba and convolution is developed to effectively extract global spatial and channel features from flood images while minimizing computational load. Finally, a spatial and channel attention module with residual connections is introduced to enable multiscale fusion and filtering of prompt information and image features across both spatial and channel dimensions, improving the utilization of prompt information. To validate the model's performance, we conduct experiments using UAV flood imagery collected from diverse regions, backgrounds, and angles. The results demonstrate that, under consistent prompt conditions, our model extracts flood areas more efficiently, reducing misclassification and omission errors. Compared with the next best benchmark model, the intersection over union for the flood category improves by at least 3.75%.