YOLO-CBNet: A robust attention-enhanced detection framework for underwater fish recognition in aquaculture environments
Mahdi Hamzaoui, Leila Bousbia, Mohamed Ould-Elhassen Aoueileyine, Imen Filali, Ridha Bouallegue · 2026
Read YOLO-CBNet: A robust attention-enhanced detection framework for underwater fish recognition in aquaculture environments by Mahdi Hamzaoui, Leila Bousbia, Mohamed Ould-Elhassen Aoueileyine, Imen Filali, Ridha Bouallegue (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0341525.txt.
Description
Mahdi Hamzaoui led a team of five authors building vision AI for fish farms, published in PLOS ONE in 2026. Fish farms need reliable eyes underwater, where light is poor and pens are crowded. Their framework, YOLO-CBNet, adds attention to a fast detection model to recognize fish in aquaculture settings.
Hamzaoui's model keeps working in murky water and dense schools, where ordinary vision systems give up. Reliable fish detection supports counting, health checks, and feeding control, all of which decide a farm's profit and the fish's welfare. Attention helps the model focus on fish rather than nets, shadows, and debris. Farms produce a growing share of the world's fish, so tools that work in real pen conditions matter well beyond one lab.
Catalog record
| CREATOR | Mahdi Hamzaoui, Leila Bousbia, Mohamed Ould-Elhassen Aoueileyine, Imen Filali, Ridha Bouallegue |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 59388 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |