Explainable hybrid multi-branch CNN–ViT–GNN framework for robust hibiscus leaf disease classification
Md Masum Billah, Saifuddin Sagor, Shahariar Hossain, Tahani Jaser Alahmadi, Mohammad Ali Moni, Mohammad Shorif Uddin · 2026
Read Explainable hybrid multi-branch CNN–ViT–GNN framework for robust hibiscus leaf disease classification by Md Masum Billah, Saifuddin Sagor, Shahariar Hossain, Tahani Jaser Alahmadi, Mohammad Ali Moni, Mohammad Shorif Uddin (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/PMC13507137.txt.
Description
Md Masum Billah leads a 2026 team that built an explainable AI system to classify hibiscus leaf diseases from field photos. Field images suffer from clutter, shifting light, and scattered spots that break simple pattern rules. Their system blends three branches: one for fine texture, one for whole-leaf context, and one for links between distant spots.
Training used a curated set of 1165 images from Bangladesh, expanded to 8000 for robust learning. Accuracy reached over 98 percent on hibiscus tests, with strong scores on two outside plant datasets and fast answers in tens of milliseconds. Four explanation tools confirmed that choices rested on real symptom patterns, not background clues.
Catalog record
| CREATOR | Md Masum Billah, Saifuddin Sagor, Shahariar Hossain, Tahani Jaser Alahmadi, Mohammad Ali Moni, Mohammad Shorif Uddin |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 91103 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |