DPC-YOLO: Underwater Plastic Pollution Detection Using YOLOv11 with Dark Prior Channel Contrast Enhancement

Underwater plastic pollution harms marine ecosystems, yet detection remains challenging due to poor visibility and color distortion in underwater images. This research offers DPCYOLO, an enhanced detection framework integrating YOLOv11 with a Dark Prior Channel Contrast Enhancement (DPCCE) technique to improve image clarity and contrast before detection. The approach offers better feature extraction under demanding underwater settings. Experimental results demonstrate that DPCYOLO obtains a precision of 98.01%, recall of 97.50%, F1-score of 98.59%, and mAP of 97.00%, indicating strong accuracy and resilience for real-time underwater plastic debris identification and marine pollution monitoring.