Deep learning model detects 'dark vessels' using satellite nightlight data

Researchers have developed a deep learning model, YOLO11, to detect fishing vessels using nighttime light satellite imagery. This dual-branch architecture, processing both panchromatic and RGB data, achieved high performance metrics including a 0.96 mAP@50. When applied to the coast of India, the model identified over 31,000 vessel instances, with a significant majority (77.3%) identified as potential "dark vessels" lacking AIS transmission. The study also revealed peak fishing activity between January and April, concentrated within 50-100 km of the coastline. AI

IMPACT Enhances maritime surveillance capabilities and provides insights into fishing patterns and regulatory compliance.

RANK_REASON Academic paper detailing a new deep learning model and its application. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• India
• Prasun Kumar Gupta
• YOLO11
• YOLO11s
• Yolov5s
• YOLOv8s

• paper
• product
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →