Related Experiment Video
Updated: Apr 8, 2026
09:15
DNA-based Fish Species Identification Protocol
Published on: April 28, 2010
29.6K
A Method for Recognizing Dead Sea Bass Based on Improved YOLOv8n.
Lizhen Zhang1, Chong Xu1, Sai Jiang1
1College of Engineering Science and Technology, Shanghai Ocean University, Shanghai 201306, China.
Sensors (Basel, Switzerland)
|July 30, 2025
Summary
Prompt detection of dead sea bass is crucial for aquaculture. This study introduces YOLOv8n-Deadfish, a lightweight model improving dead fish detection accuracy and speed while reducing computational costs.
Area of Science:
- Aquaculture technology
- Computer vision
- Machine learning
Background:
- Dead fish in aquaculture lead to pollution and disease spread.
- Existing object detection models for dead fish are complex and inaccurate, especially with occlusion.
Purpose of the Study:
- To develop a lightweight, high-precision object detection model for identifying dead sea bass.
- To improve the efficiency and accuracy of dead fish detection in intelligent aquaculture systems.
Main Methods:
- Developed YOLOv8n-Deadfish, incorporating an expanded dataset and a novel C2f-faster-EMA module.
- Integrated a weighted bidirectional feature pyramid network (BiFPN) for enhanced feature fusion.
- Utilized the Inner-CIoU loss function to improve bounding box regression and model convergence.
Main Results:
- YOLOv8n-Deadfish achieved 90.0% accuracy, 90.4% recall, and 93.6% mean precision.
- Model parameters and GFLOPs were reduced by 23.3% and 18.5%, respectively.
- Detection speed increased to 424.6 FPS, a significant improvement over the base YOLOv8n.
Conclusions:
- The YOLOv8n-Deadfish model offers a superior solution for detecting dead sea bass in aquaculture.
- This technology provides a foundation for more effective and efficient intelligent aquaculture management.
- The model's lightweight design and high performance address limitations of previous detection methods.
