Enhanced hermit crabs detection using super-resolution reconstruction and improved YOLOv8 on UAV-captured imagery

Hermit crabs are vital to coastal ecosystems, serving as environmental health indicators and contributing to seed dispersal, debris cleanup, and soil disturbance. Traditional hermit crabs survey methods, like quadrat sampling, are labor-intensive and environmentally dependent. This study presents an innovative approach combining UAV (Unmanned Aerial Vehicles)-based remote sensing with Super-Resolution Reconstruction (SRR) and the CRAB-YOLO detection network, a modification of YOLOv8s, to monitor hermit crabs effectively. SRR enhances image quality by addressing motion blur and insufficient resolution, significantly improving detection accuracy over conventional low-resolution fuzzy images. CRAB-YOLO integrates three improvements for accuracy, hermit crab characteristics, and computational efficiency, achieving state-of-the-art (SOTA) performance. The Residual Dense Network (RDN) demonstrated the best image reconstruction performance, and CRAB-YOLO achieved a mean average precision (mAP) of 69.5 % on the SRR test set, a 40 % improvement over the conventional Bicubic method with a magnification factor of 4. These results prove the effectiveness of the proposed method for cost-effective, automated hermit crab monitoring supporting coastal benthos conservation efforts.