A deep learning-based fusion framework for robust fine-grained classification of sea turtles in support of marine biodiversity
Parkpoom Chaisiriprasert, Apicha Deearom · 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 (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-0344942.txt.
Description
Parkpoom Chaisiriprasert and Apicha Deearom built a learning system to tell sea turtle species apart in underwater photos. Telling species apart supports counts and protection, yet murky water, poor light, and blocked views fool both people and models. Their system feeds both colour photos and edge maps that trace shell and flipper outlines into a shared network with smart blending steps.
The best blending version led on precision, recall, and overall detection scores, beating simpler fusions. Adding outline cues helped most where species look alike in colour alone. Automated, accurate sorting could let survey teams process far more dive and drone photos, building better maps of where each turtle species lives.
Catalog record
| CREATOR | Parkpoom Chaisiriprasert, Apicha Deearom |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 87423 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |