Related Experiment Video
Updated: Jul 7, 2026

VisioTracker, an Innovative Automated Approach to Oculomotor Analysis
Published on: October 12, 2011
An Improved YOLOv8n Used for Fish Detection in Natural Water Environments
Zehao Zhang1,2,3,4, Yi Qu1,2,3,4, Tan Wang1,2,3,4
1School of Information and Artificial Intelligence, Anhui Agricultural University, Hefei 230036, China.
This study introduces BSSFISH-YOLOv8, an accurate computer vision method for fish detection in underwater environments. It enhances detection efficiency and reduces costs in fishery surveys by improving accuracy for small targets.
Area of Science:
- Fisheries Science
- Computer Vision
- Marine Biology
Background:
- Underwater photography presents challenges for accurate fish detection due to environmental complexity.
- Existing computer vision methods have limitations in accuracy for fishery resource surveys.
- Improving detection efficiency and reducing costs are crucial for effective fishery management.
Purpose of the Study:
- To develop an accurate and efficient computer vision method for fish detection in natural underwater environments.
- To address the limitations of current methods in handling underwater image complexity and small targets.
- To enhance the application of computer vision in fishery resource surveys.
Main Methods:
- Proposed BSSFISH-YOLOv8 model incorporating SPD-Conv module to preserve fine-grained information.
- Integrated BiFormer, a dynamic sparse attention technique, into the backbone network for enhanced feature focus and efficiency.
- Introduced a 160 × 160 small target detection layer (STDL) to improve sensitivity to smaller fish.
Main Results:
- BSSFISH-YOLOv8 achieved 88.3% mAP@50 and 58.3% mAP@50:95.
- Performance metrics were 2.0% and 3.3% higher than the YOLOv8n model, respectively.
- Demonstrated improved accuracy and efficiency in detecting fish in complex underwater conditions.
Conclusions:
- BSSFISH-YOLOv8 offers a significant advancement for fish detection in underwater environments.
- The method has the potential to reduce measurement costs and improve efficiency in fishery resource surveys.
- This research contributes to sustainable fishery management through enhanced technological application.
Related Concept Videos
Osmoregulation in Fishes
FISH - Fluorescent In-situ Hybridization
Imaging Biological Samples with Optical Microscopy
In optical microscopy, the specimen to be viewed is placed on a glass slide and clipped on the stage...
Quality of Water
Testing Water Quality
Freshwater Microbial Ecology

