Related Experiment Video
Updated: Jun 27, 2025

05:57
Long-term Video Tracking of Cohoused Aquatic Animals: A Case Study of the Daily Locomotor Activity of the Norway Lobster Nephrops norvegicus
Published on: April 8, 2019
6.8K
Edge computing based real-time Nephrops (Nephrops norvegicus) catch estimation in demersal trawls using object
Ercan Avsar1, Jordan P Feekings2, Ludvig Ahm Krag2
1Section for Fisheries Technology, Institute of Aquatic Resources, Technical University of Denmark, Hirtshals, Denmark. erca@aqua.dtu.dk.
Scientific Reports
|April 25, 2024
Summary
A new real-time underwater video system uses object detection to count Nephrops in demersal trawls. This technology enhances fishery sustainability and economic performance by providing immediate catch data.
Area of Science:
- Marine biology
- Fisheries science
- Computer vision
Background:
- Demersal trawl fisheries face challenges with delayed catch data, impacting sustainability and economics.
- In-trawl cameras offer real-time catch observation potential.
- Processing this data is crucial for improving fishery management.
Purpose of the Study:
- To develop and evaluate a real-time underwater video processing system for counting Nephrops individuals.
- To optimize system performance for accurate and rapid catch assessment.
Main Methods:
- Utilized object detection and tracking on an edge device (NVIDIA Jetson AGX Orin).
- Tested seven state-of-the-art YOLO models and evaluated four frame skipping techniques.
- Focused on achieving simultaneous real-time processing and accurate counting.
Main Results:
- The YOLOv8s model combined with adaptive frame skipping achieved 97.47 FPS.
- Attained a correct count rate of 82.57% and an F-score of 0.86.
- Demonstrated the feasibility of real-time processing for demersal trawl fisheries.
Conclusions:
- The developed system provides real-time catch information for Nephrops-directed fisheries.
- This technology can significantly improve the sustainability and economic viability of these fisheries.
- Real-time data processing is key to addressing drawbacks in traditional trawl fisheries.

