Detecting beach litter in drone images using deep learning

Beach pollution through litter leads to various negative effects and mitigation and cleanup efforts are required to prevent accumulation. Automated monitoring is an important step towards this goal and has become much more feasible with recent developments in drone technology and artificial intelligence / object detection. To assess the potential of artificial intelligence for litter monitoring on Maltese beaches, two deep learning algorithms (YOLOv5 and Faster R-CNN) were trained on drone footage of beach litter and their detection performance compared. With a mean average precision (mAP) of 0.542 YOLOv5 outperformed Faster R-CNN (mAP: 0.328). In addition, detected litter objects were geolocated so that their position could be estimated with an average error of 3.7 meters.

Automated summary

Beach pollution from litter has negative effects that require mitigation and cleanup efforts. Automated monitoring using drone technology and artificial intelligence/object detection has become more feasible. In this study, two deep learning algorithms (YOLOv5 and Faster R-CNN) were trained on drone footage to monitor litter on Maltese beaches. YOLOv5 outperformed Faster R-CNN in terms of detection performance, with an average precision (mAP) of 0.542 compared to 0.328. The geolocation of detected litter objects had an average estimation error of 3.7 meters.