Detection of Endangered Deer Species Using UAV Imagery: A Comparative Study Between Efficient Deep Learning Approaches

TL;DR

This paper compares advanced neural network models, including YOLOv11 and RT-DETR, for detecting endangered marsh deer in UAV images, emphasizing segmentation for improved accuracy in challenging scenarios.

Contribution

It introduces a segmentation-augmented YOLO model and provides a comparative analysis of neural networks for wildlife detection in UAV imagery.

Findings

Segmentation-enhanced YOLO outperforms other models in detection accuracy.

Incorporating segmentation masks improves detection of occluded and small deer.

The study offers insights for scalable UAV-based wildlife monitoring.

Abstract

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.