Development of a ship-based camera monitoring system for floating marine debris

• Published in: Marine pollution bulletin 2024-09, Vol.206, p.116722, Article 116722
• Main Authors: Yang, Ruofei, Uchida, Keiichi, Miyamoto, Yoshinori, Arakawa, Hisayuki, Hagita, Ryuichi, Aikawa, Tetsutaro
• Format: Article
• Language: English
• Subjects: algorithms; Automatic monitoring system; cameras; Deep learning; Floating marine debris; marine debris; Marine pollution; Visual survey; YOLOv8
• ISSN: 0025-326X; 1879-3363; 1879-3363
• DOI: 10.1016/j.marpolbul.2024.116722

This study developed an automatic monitoring system for Floating Marine Debris (FMD) aimed at reducing the labor-intensiveness of traditional visual surveys. It involved creating a comprehensive FMD database using 55.6 h of video footage and numerous annotated images, which facilitated the training of a deep learning model based on the YOLOv8 architecture. Additionally, the study implemented the BoT-SORT algorithm for FMD tracking, significantly enhancing detection accuracy by effectively filtering out disturbances such as sea waves and seabirds, based on the movement patterns observed in FMD trajectories. Tested across 16 voyages in various marine environments, the system demonstrated high accuracy in recognizing different types of FMD, achieving a mean Average Precision (mAP@0.5) of 0.97. In terms of detecting FMD from video footage, the system reached an F1 score of 83.63 %. It showed potential as a viable substitute for manual methods for FMD larger than 20 cm. [Display omitted] •Developed a prototype system for automated floating marine debris monitoring.•Introduced an automated system to assist and potentially replace visual surveys.•Utilized YOLOv8 for its rapid response and high recognition accuracy.•Filtered out factors such as sea waves and seabirds that impact monitoring.