Detection of Endangered Deer Species Using UAV Imagery: A Comparative Study Between Efficient Deep Learning Approaches
This work addresses wildlife monitoring and conservation for endangered species using UAVs, but it is incremental as it builds on existing deep learning methods with added segmentation.
This study tackled the problem of detecting endangered marsh deer in UAV imagery, where specimens are small and occluded, by comparing YOLOv11 and RT-DETR models and extending analysis with segmentation masks. The result showed that incorporating a segmentation head into a YOLO model achieved superior detection performance, though no concrete numbers were provided.
This study compares the performance of state-of-the-art neural networks including variants of the YOLOv11 and RT-DETR models for detecting marsh deer in UAV imagery, in scenarios where specimens occupy a very small portion of the image and are occluded by vegetation. We extend previous analysis adding precise segmentation masks for our datasets enabling a fine-grained training of a YOLO model with a segmentation head included. Experimental results show the effectiveness of incorporating the segmentation head achieving superior detection performance. This work contributes valuable insights for improving UAV-based wildlife monitoring and conservation strategies through scalable and accurate AI-driven detection systems.