A deep learning-based fusion framework for robust fine-grained classification of sea turtles in support of marine biodiversity
Parkpoom Chaisiriprasert, Apicha Deearom, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla · 2026
Read A deep learning-based fusion framework for robust fine-grained classification of sea turtles in support of marine biodiversity by Parkpoom Chaisiriprasert, Apicha Deearom, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla (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/pmc-13249215.txt.
Description
Chaisiriprasert and Deearom taught a computer to name sea turtle species from underwater photos, in a 2026 study. Murky water, poor light, and look-alike species make human sorting slow and error-prone. Their system feeds a standard image network both color pictures and edge maps that trace shell and flipper outlines, then merges the two streams with attention modules they designed.
The stronger of their two fusion designs beat the baseline on every score, reaching the best precision, recall, and overall detection accuracy. Edge information helped most where species look nearly identical in color alone. The authors present the framework as a scalable tool for reef monitoring and wider marine biodiversity surveys.
Catalog record
| CREATOR | Parkpoom Chaisiriprasert, Apicha Deearom, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla, Yaseen Ahmed Al-Mulla |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 138 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |