Although advances in deep learning and aerial surveillance technology are improving wildlife conservation efforts, complex and erratic environmental conditions still pose a problem, requiring innovative solutions for cost-effective small animal detection. This work introduces DEAL-YOLO, a novel approach that improves small object detection in Unmanned Aerial Vehicle (UAV) images by using multi-objective loss functions like Wise IoU (WIoU) and Normalized Wasserstein Distance (NWD), which prioritize pixels near the centre of the bounding box, ensuring smoother localization and reducing abrupt deviations. Additionally, the model is optimized through efficient feature extraction with Linear Deformable (LD) convolutions, enhancing accuracy while maintaining computational efficiency. The Scaled Sequence Feature Fusion (SSFF) module enhances object detection by effectively capturing inter-scale relationships, improving feature representation, and boosting metrics through optimized multiscale fusion. Comparison with baseline models reveals high efficacy with up to 69.5\% fewer parameters compared to vanilla Yolov8-N, highlighting the robustness of the proposed modifications. Through this approach, our paper aims to facilitate the detection of endangered species, animal population analysis, habitat monitoring, biodiversity research, and various other applications that enrich wildlife conservation efforts. DEAL-YOLO employs a two-stage inference paradigm for object detection, refining selected regions to improve localization and confidence. This approach enhances performance, especially for small instances with low objectness scores.
Performance Optimization of YOLO-FEDER FusionNet for Robust Drone Detection in Visually Complex Environments
Institute for the Protection of Terrestrial Infrastructures, German Aerospace Center（地面基础设施保护研究所，德国航天中心）；ACIDA Lab, University of Applied Sciences Darmstadt（达姆施塔特应用科学大学ACIDA实验室）；European University of Technology（欧洲技术大学）
WildLIFT: Lifting monocular drone video to 3D for species-agnostic wildlife monitoring
Fondazione Bruno Kessler（布雷诺·科塞勒基金会）；University of Muenster（明斯特大学）；Max Planck Institute of Animal Behaviour（动物行为马克斯·普朗克研究所）；University of Konstanz（康斯坦茨大学）
LAF-YOLOv10 with Partial Convolution Backbone, Attention-Guided Feature Pyramid, Auxiliary P2 Head, and Wise-IoU Loss for Small Object Detection in Drone Aerial Imagery
Department of Information Technology（信息科技系）；Department of Computer Science and Engineering（计算机科学与工程系）
如今，无人机等无人机被广泛用于各种目的，如从ariel图像中捕获和目标检测等。这些小型ariel车辆容易进入公众可能会造成严重的安全威胁。例如，关键的地方可能会被间谍使用无人机混在公共场所进行监视。Study in Hand提出了一种改进且高效的基于深度学习的自主系统，可以非常精确地检测和跟踪非常小的无人机。该系统由一个自定义的深度学习模型Tiny YOLOv3组成，这是一种非常快速的目标检测模型You Look Only Once（YOLO），它被构建并用于检测。目标检测算法将有效地检测无人机。与之前的YOLO版本相比，所提出的架构表现出了明显更好的性能。在资源使用和时间复杂度方面观察到改进。性能是使用召回率和精确率分别为93%和91%的指标来衡量的。
YOLOBirDrone: Dataset for Bird vs Drone Detection and Classification and a YOLO based enhanced learning architecture
Department of Intelligent Machine and Computing System, CSIR CSIO（智能机器与计算系统部门、CSIR CSIO）；School of Computing and Electrical Engineering, IIT Mandi（计算与电子工程学院、IIT Mandi）
Improved YOLOv5 Based on Attention Mechanism and FasterNet for Foreign Object Detection on Railway and Airway tracks
Stevens Institute of Technology（史蒂文斯理工学院）；Northern Arizona University（北亚利桑那大学）；University of Electronic Science and Technology of China（电子科技大学）；Shenzhen SmartChip Microelectronics Technology Co., Ltd.（深圳市智芯微电子技术有限公司）
Automatic detection of aerial survey ground control points based on Yolov5-OBB
Henan University of Urban Construction（河南城建学院）；University of North Carolina（北卡罗来纳大学）；Northeast Institute of Geography and Agricultural Ecology, Chinese Academy of Sciences（中国科学院东北地理与农业生态研究所）；Zhengzhou University（郑州大学）
Deep Learning Object Detection Methods for Ecological Camera Trap Data
University of Guelph（圭尔夫大学）；Vector Institute for Artificial Intelligence（向量人工智能研究所）；Canadian Institute for Advanced Research（加拿大高级研究所）
用于计算机视觉任务的深度学习方法有望实现相机陷阱图像数据分析的自动化。生态相机陷阱是监测生态系统动物种群的常用方法，因为它们提供对环境的持续洞察而不会侵入。然而，相机陷阱图像的分析是昂贵的，劳动密集型的，并且耗时。用于目标检测的深度学习领域的最新进展显示出自动分析相机陷阱图像的前景。在这里，我们通过训练和比较两个深度学习目标检测分类器Faster R-CNN和YOLO v2.0来展示它们的能力，以使用Reconyx Camera Trap和自标记的Gold Standard Snapshot Serengeti数据集识别，量化和定位相机陷阱图像中的动物物种。当在大型标记数据集上训练时，对象识别方法已经显示出成功。我们展示了他们的使用，在现实规模的生态数据集的背景下，通过测试，如果目标检测方法是适用于生态研究的情况下，利用迁移学习。更快的R-CNN优于YOLO v2.0，在两个数据集上的平均准确率分别为93.0%和76.7%。我们的研究结果显示，在标记相机陷阱图像的繁重任务的自动化方面迈出了有希望的步伐，这可用于提高我们对全球生态系统种群动态的理解。
Towards automatic smoke detector inspection: Recognition of the smoke detectors in industrial facilities and preparation for future drone integration
Brno University of Technology（布尔诺理工大学）；VSB - Technical University of Ostrava（俄斯特拉发技术大学）；University of Ostrava（俄斯特拉发大学）；Mendel University in Brno（布尔诺孟德尔大学）
USU-Corn-WeedDB: A UAV RGB Image Dataset for Multi-Species Weed Detection in Forage Corn
Department of Applied Sciences, Technology, and Education, Utah State University（应用科学、技术和教育系，犹他州立大学）；Department of Plants, Soils & Climate, Utah State University（植物、土壤与气候系，犹他州立大学）；Department of Civil and Environmental Engineering, Utah State University（土木与环境工程系，犹他州立大学）
