Search Results (1,305)

Unmanned Aerial Vehicles (UAVs) have rapidly grown into different essential applications, including surveillance, disaster response, agriculture, and urban monitoring. However, for UAVS to steer safely and autonomously, the ability to detect obstacles and nearby aircraft remains crucial, especially under hard environmental conditions. This study comprehensively analyzes the recent landscape of obstacle and aircraft detection techniques tailored for UAVs acting in difficult scenarios such as fog, rain, smoke, low light, motion blur, and disorderly environments. It starts with a detailed discussion of key detection challenges and continues with an evaluation of different sensor types, from RGB and infrared cameras to LiDAR, radar, sonar, and event-based vision sensors. Both classical computer vision methods and deep learning-based detection techniques are examined in particular, highlighting their performance strengths and limitations under degraded sensing conditions. The paper additionally offers an overview of suitable UAV-specific datasets and the evaluation metrics generally used to evaluate detection systems. Finally, the paper examines open problems and coming research directions, emphasising the demand for lightweight, adaptive, and weather-resilient detection systems appropriate for real-time onboard processing. This study aims to guide students and engineers towards developing stronger and intelligent detection systems for next-generation UAV operations. Full article

Vehicle re-identification (Re-ID) is a critical computer vision task that aims to match the same vehicle across spatially distributed cameras, especially in the context of remote sensing imagery. While prior research has primarily focused on Re-ID using remote sensing images captured from similar, typically elevated viewpoints, these settings do not fully reflect complex aerial-ground collaborative remote sensing scenarios. In this work, we introduce a novel and challenging task: aerial-ground cross-view vehicle Re-ID, which involves retrieving vehicles in ground-view image galleries using query images captured from aerial (top-down) perspectives. This task is increasingly relevant due to the integration of drone-based surveillance and ground-level monitoring in multi-source remote sensing systems, yet it poses substantial challenges due to significant appearance variations between aerial and ground views. To support this task, we present AGID (Aerial-Ground Vehicle Re-Identification), the first benchmark dataset specifically designed for aerial-ground cross-view vehicle Re-ID. AGID comprises 20,785 remote sensing images of 834 vehicle identities, collected using drones and fixed ground cameras. We further propose a novel method, Enhanced Self-Correlation Feature Computation (ESFC), which enhances spatial relationships between semantically similar regions and incorporates shape information to improve feature discrimination. Extensive experiments on the AGID dataset and three widely used vehicle Re-ID benchmarks validate the effectiveness of our method, which achieves a Rank-1 accuracy of 69.0% on AGID, surpassing state-of-the-art approaches by 2.1%. Full article

In this study, we introduce the application of deep learning (DL) models, specifically convolutional neural networks (CNNs), for detecting the health status of date palm leaves using images captured by an unmanned aerial vehicle (UAV). The images are modeled using the Newton–Euler method to ensure stability and accurate image acquisition. These deep learning models are implemented by a voting-based classification (VBC) system that combines multiple CNN architectures, including MobileNet, a handcrafted CNN, VGG16, and VGG19, to enhance classification accuracy and robustness. The classifiers independently generate predictions, and a voting mechanism determines the final classification. This hybridization of image-based visual servoing (IBVS) and classifiers makes immediate adaptations to changing conditions, providing straightforward and smooth flying as well as vision classification. The dataset used in this study was collected using a dual-camera UAV, which captures high-resolution images to detect pests in date palm leaves. After applying the proposed classification strategy, the implemented voting method achieved an impressive accuracy of 99.16% on the test set for detecting health conditions in date palm leaves, surpassing individual classifiers. The obtained results are discussed and compared to show the effectiveness of this classification technique. Full article

Low-altitude drones, which are unimpeded by traffic congestion or urban terrain, have become a critical asset in emergency rescue missions. To address the current lack of emergency rescue data, UAV aerial videos were collected to create an experimental dataset for action classification and localization annotation. A total of 5082 keyframes were labeled with 1–5 targets each, and 14,412 instances of data were prepared (including flight altitude and camera angles) for action classification and position annotation. To mitigate the challenges posed by high-resolution drone footage with excessive redundant information, we propose the SlowFast-Traffic (SF-T) framework, a spatio-temporal sequence-based algorithm for recognizing traffic accident rescue actions. For more efficient extraction of target–background correlation features, we introduce the Actor-Centric Relation Network (ACRN) module, which employs temporal max pooling to enhance the time-dimensional features of static backgrounds, significantly reducing redundancy-induced interference. Additionally, smaller ROI feature map outputs are adopted to boost computational speed. To tackle class imbalance in incident samples, we integrate a Class-Balanced Focal Loss (CB-Focal Loss) function, effectively resolving rare-action recognition in specific rescue scenarios. We replace the original Faster R-CNN with YOLOX-s to improve the target detection rate. On our proposed dataset, the SF-T model achieves a mean average precision (mAP) of 83.9%, which is 8.5% higher than that of the standard SlowFast architecture while maintaining a processing speed of 34.9 tasks/s. Both accuracy-related metrics and computational efficiency are substantially improved. The proposed method demonstrates strong robustness and real-time analysis capabilities for modern traffic rescue action recognition. Full article

Precision irrigation plays a crucial role in managing crop production in a sustainable and environmentally friendly manner. This study builds on the results of the GreenWaterDrone project, aiming to estimate, in real time, the actual water requirements of crop fields using the crop water stress index, integrating infrared canopy temperature, air temperature, relative humidity, and thermal and near-infrared imagery. To achieve this, a state-of-the-art aerial micrometeorological station (AMMS), equipped with an infrared thermal sensor, temperature–humidity sensor, and advanced multispectral and thermal cameras is mounted on an unmanned aerial system (UAS), thus minimizing crop field intervention and permanently installed equipment maintenance. Additionally, data from satellite systems and ground micrometeorological stations (GMMS) are integrated to enhance and upscale system results from the local field to the regional level. The research was conducted over two years of pilot testing in the municipality of Trifilia (Peloponnese, Greece) on pilot potato and watermelon crops, which are primary cultivations in the region. Results revealed that empirical irrigation applied to the rhizosphere significantly exceeded crop water needs, with over-irrigation exceeding by 390% the maximum requirement in the case of potato. Furthermore, correlations between high-resolution remote and proximal sensors were strong, while associations with coarser Landsat 8 satellite data, to upscale the local pilot field experimental results, were moderate. By applying a comprehensive model for upscaling pilot field results, to the overall Trifilia region, project findings proved adequate for supporting sustainable irrigation planning through simulation scenarios. The results of this study, in the context of the overall services introduced by the project, provide valuable insights for farmers, agricultural scientists, and local/regional authorities and stakeholders, facilitating improved regional water management and sustainable agricultural policies. Full article

This work aims to assess the droplets produced by a novel evaporation process, proposed as an alternative for managing tequila vinasses, using a spectral camera with three spectral bands and a thermal camera mounted on an unmanned aerial vehicle (UAV). High-resolution satellite images with seven spectral bands complemented this characterization. The spectral characterization was conducted by comparing three experimental conditions: the background of the study area without droplets, the droplets generated from purified water, and the droplets produced from tequila vinasses. Two monitoring campaigns, conducted in November 2024 and January 2025, revealed that the tequila vinasse droplets exhibited a maximum influence radius of 16 m, primarily regulated by wind speed conditions (6–16 km/h). Thermal analysis identified the droplet plume as a zone with a lower temperature, creating a thermal contrast of up to 6.6 °C against the average background temperature of 36.6 °C. No significant difference was observed in the influence radius between the droplets generated from vinasse and those from potable water. Spectral analysis of the UAV and satellite images showed significant (p < 0.05) differences in reflectance when the droplets were present (e.g., the coastal blue band increased from an average of 14.43 to 95.59 when vinasse droplets were present). This suggests that the presence of chemical compounds altered light absorption and reflection. However, the instrument’s sensitivity limited the detection of organic compounds at concentrations below its detection limit. The monitoring data presented in this manuscript is crucial for developing strategies to mitigate the potential environmental impacts of the droplets emitted by this novel process. Full article

Transpiration is the dominant process driving water loss in crops, significantly influencing their growth, development, and yield. Efficient monitoring of transpiration rate (Tr) is crucial for evaluating crop physiological status and optimizing water management strategies. The three-temperature (3T) model has potential for rapid estimation of transpiration rates, but its application to low-altitude remote sensing has not yet been further investigated. To evaluate the performance of 3T model based on land surface temperature (LST) and canopy temperature (T_C) in estimating transpiration rate, this study utilized an unmanned aerial vehicle (UAV) equipped with a thermal infrared (TIR) camera to capture TIR images of summer maize during the nodulation-irrigation stage under four different moisture treatments, from which LST was extracted. The Gaussian Hidden Markov Random Field (GHMRF) model was applied to segment the TIR images, facilitating the extraction of T_C. Finally, an improved 3T model incorporating fractional vegetation coverage (FVC) was proposed. The findings of the study demonstrate that: (1) The GHMRF model offers an effective approach for TIR image segmentation. The mechanism of thermal TIR segmentation implemented by the GHMRF model is explored. The results indicate that when the potential energy function parameter β value is 0.1, the optimal performance is provided. (2) The feasibility of utilizing UAV-based TIR remote sensing in conjunction with the 3T model for estimating Tr has been demonstrated, showing a significant correlation between the measured and the estimated transpiration rate (T_r-3T_C), derived from T_C data obtained through the segmentation and processing of TIR imagery. The correlation coefficients (r) were 0.946 in 2022 and 0.872 in 2023. (3) The improved 3T model has demonstrated its ability to enhance the estimation accuracy of crop Tr rapidly and effectively, exhibiting a robust correlation with T_r-3T_C. The correlation coefficients for the two observed years are 0.991 and 0.989, respectively, while the model maintains low RMSE of 0.756 mmol H₂O m⁻² s⁻¹ and 0.555 mmol H₂O m⁻² s⁻¹ for the respective years, indicating strong interannual stability. Full article

Innovative applications of high-resolution airborne imaging are explored for detecting grapevine diseases. Driven by the motivation to enhance early disease detection, the method’s effectiveness lies in its capacity to identify isolated cases of grapevine yellows (Flavescence dorée and Bois Noir) and trunk disease (Esca complex), crucial for preventing the disease from spreading to unaffected areas. Conducted over a 17 ha vineyard in the Forlì municipality in Emilia-Romagna (Italy), the aerial survey utilized a photogrammetric camera capturing centimeter-level resolution images of the whole area in 17 minutes. These images were then processed through an automated analysis leveraging RGB-based spectral indices (Green–Red Vegetation Index—GRVI, Green–Blue Vegetation Index—GBVI, and Blue–Red Vegetation Index—BRVI). The analysis scanned the 1.24 · 10⁹ pixels of the orthomosaic, detecting 0.4% of the vineyard area showing evidence of disease. The instances, density, and incidence maps provide insights into symptoms’ spatial distribution and facilitate precise interventions. High specificity (0.96) and good sensitivity (0.56) emerged from the ground field observation campaign. Statistical analysis revealed a significant edge effect in symptom distribution, with higher disease occurrence near vineyard borders. This pattern, confirmed by spatial autocorrelation and non-parametric tests, likely reflects increased vector activity and environmental stress at the vineyard margins. The presented pilot study not only provides a reliable detection tool for grapevine diseases but also lays the groundwork for an early warning system that, if extended to larger areas, could offer a valuable system to guide on-the-ground monitoring and facilitate strategic decision-making by the authorities. Full article

This paper presents FWStab, a specialized video stabilization dataset tailored for flapping-wing platforms. The dataset encompasses five typical flight scenarios, featuring 48 video clips with intense dynamic jitter. The corresponding Inertial Measurement Unit (IMU) sensor data are synchronously collected, which jointly provide reliable support for multimodal modeling. Based on this, to address the issue of poor image acquisition quality due to severe vibrations in aerial vehicles, this paper proposes a multi-modal signal fusion video stabilization framework. This framework effectively integrates image features and inertial sensor features to predict smooth and stable camera poses. During the video stabilization process, the true camera motion originally estimated based on sensors is warped to the smooth trajectory predicted by the network, thereby optimizing the inter-frame stability. This approach maintains the global rigidity of scene motion, avoids visual artifacts caused by traditional dense optical flow-based spatiotemporal warping, and rectifies rolling shutter-induced distortions. Furthermore, the network is trained in an unsupervised manner by leveraging a joint loss function that integrates camera pose smoothness and optical flow residuals. When coupled with a multi-stage training strategy, this framework demonstrates remarkable stabilization adaptability across a wide range of scenarios. The entire framework employs Long Short-Term Memory (LSTM) to model the temporal characteristics of camera trajectories, enabling high-precision prediction of smooth trajectories. Full article

Camera-centric perception has matured into a cornerstone of modern autonomy, from self-driving cars and factory cobots to underwater and planetary exploration. This review synthesizes more than a decade of progress in vision-based robotic navigation through an engineering lens, charting the full pipeline from sensing to deployment. We first examine the expanding sensor palette—monocular and multi-camera rigs, stereo and RGB-D devices, LiDAR–camera hybrids, event cameras, and infrared systems—highlighting the complementary operating envelopes and the rise of learning-based depth inference. The advances in visual localization and mapping are then analyzed, contrasting sparse and dense SLAM approaches, as well as monocular, stereo, and visual–inertial formulations. Additional topics include loop closure, semantic mapping, and LiDAR–visual–inertial fusion, which enables drift-free operation in dynamic environments. Building on these foundations, we review the navigation and control strategies, spanning classical planning, reinforcement and imitation learning, hybrid topological–metric memories, and emerging visual language guidance. Application case studies—autonomous driving, industrial manipulation, autonomous underwater vehicles, planetary rovers, aerial drones, and humanoids—demonstrate how tailored sensor suites and algorithms meet domain-specific constraints. Finally, the future research trajectories are distilled: generative AI for synthetic training data and scene completion; high-density 3D perception with solid-state LiDAR and neural implicit representations; event-based vision for ultra-fast control; and human-centric autonomy in next-generation robots. By providing a unified taxonomy, a comparative analysis, and engineering guidelines, this review aims to inform researchers and practitioners designing robust, scalable, vision-driven robotic systems. Full article

This study presents a cooperative unmanned aerial system (UAS) designed to enable precise autonomous landings in unstructured environments using low-cost onboard vision technology. This approach involves a carrier UAV with a stabilized RGB camera and a neural inference system, as well as a lightweight tailsitter payload UAV with an embedded grayscale vision module. The system relies on visually recognizable landing markers and does not require additional sensors. Field trials comprising full deployments achieved an 80% success rate in autonomous landings, with vertical touchdown occurring within a 1.5 m radius of the target. These results confirm that vision-based marker detection using compact neural models can effectively support autonomous UAV operations in constrained conditions. This architecture offers a scalable alternative to the high complexity of SLAM or terrain-mapping systems. Full article

Semantic segmentation using neural networks (NNs) has significant potential for weed detection in agricultural fields. However, conventional datasets captured from aerial perspectives often fail to detect weeds that are either hidden beneath crops or submerged in water. This study proposes a method for accurately detecting weed pixels through ensemble learning-based semantic segmentation, using forward-facing images captured by a camera mounted on an aquatic drone navigating between rice plants. We also present a paddy field weed image dataset constructed to train the NN models. Multiple semantic segmentation models were trained, compared, and evaluated, achieving a weed intersection over union (IoU) of 0.441, mean IoU (mIoU) of 0.706, and pixel accuracy of 0.971. Full article

This work proposes an end-to-end solution for leaf segmentation, disease detection, and damage quantification, specifically focusing on citrus crops. The primary motivation behind this research is to enable the early detection of phytosanitary problems, which directly impact the productivity and profitability of Spanish and Portuguese agricultural developments, while ensuring environmentally safe management practices. It integrates an onboard computing module for Unmanned Aerial Vehicles (UAVs) using a Raspberry Pi 4 with Global Positioning System (GPS) and camera modules, allowing the real-time geolocation of images in citrus croplands. To address the lack of public data, a comprehensive database was created and manually labelled at the pixel level to provide accurate training data for a deep learning approach. To reduce annotation effort, we developed a custom automation algorithm for pixel-wise labelling in complex natural backgrounds. A SegNet architecture with a Visual Geometry Group 16 (VGG16) backbone was trained for the semantic, pixel-wise segmentation of citrus foliage. The model was successfully integrated as a modular component within a broader system architecture and was tested with UAV-acquired images, demonstrating accurate disease detection and quantification, even under varied conditions. The developed system provides a robust tool for the efficient monitoring of citrus crops in precision agriculture. Full article

The autonomous landing of unmanned aerial vehicles (UAVs) on moving platforms has potential applications across various domains. However, robust landing remains challenging because the detection reliability of UAVs decreases when the UAV is close to a moving platform. To address this issue, this paper proposes a novel landing strategy that ensures a high detection rate. First, a robust detectable region was established by considering the sensing range and maneuverability limitations of the UAV. Second, a vector field was designed to guide the UAV to the moving platform while remaining in a robust detectable region. Next, safe and accurate landings were achieved by considering the current velocity and vector field. The landing strategy was validated through outdoor flight experiments. A quadrotor equipped with a gimbal-mounted camera was used, and a fractal marker was attached to the moving platform for detection and tracking. When the moving platform moved at a speed of 2–4.3 m/s, the UAV successfully landed on the platform with a distance error of 0.4 m. Because of the robust detectable region and vector field, the detection was conducted with a high success rate (94.9%). Full article

The rapid development of unmanned aerial vehicles (UAVs) has had a significant impact on the growth of the economic, industrial, and social welfare of society. The possibility of reaching places that are difficult and dangerous for humans to access with minimal use of third-party resources increases the efficiency and quality of maintenance of construction structures, agriculture, and exploration, which are carried out with the help of drones with a predetermined trajectory. The widespread use of UAVs has caused problems with the control of the drones’ correctness following a given route, which leads to emergencies and accidents. Therefore, UAV monitoring with video cameras is of great importance. In this paper, we propose a Yolov12 architecture with positive–negative pulse-based optimization algorithms to solve the problem of drone detection on video data. Self-attention-based mechanisms in transformer neural networks (NNs) improved the quality of drone detection on video. The developed algorithms for training NN architectures improved the accuracy of drone detection by achieving the global extremum of the loss function in fewer epochs using positive–negative pulse-based optimization algorithms. The proposed approach improved object detection accuracy by 2.8 percentage points compared to known state-of-the-art analogs. Full article