A two-stage deep-learning framework for image classification of 46 southeast Asian fruit and vegetable categories

Decho Surangsrirat, Panyawut Sri-iesaranusorn, Warisara Asawaponwiput, Wassapon Watanakeesuntorn, Polathep Vichitkunakorn · 2026

Read A two-stage deep-learning framework for image classification of 46 southeast Asian fruit and vegetable categories by Decho Surangsrirat, Panyawut Sri-iesaranusorn, Warisara Asawaponwiput, Wassapon Watanakeesuntorn, Polathep Vichitkunakorn (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-13550162.txt.

Description

Decho Surangsrirat and four co-authors build an image classifier for Southeast Asian fruits and vegetables in this 2026 paper. The work supports ThaiSook, a health app that tracks food intake in Thailand. Staff had been sorting user food photos by hand, a slow and error prone task. The new system sorts photos into forty-six food classes on its own.

The system works in two stages. The first stage screens out photos that show no fruit or vegetable at all. The second stage names the exact food, using a model trained with extra unlabeled photos from app users. End to end tests gave top scores above ninety-two percent for the first guess. The authors say the tool is robust enough for use on phones with modest power, where it can speed dietary tracking.

Catalog record

CREATORDecho Surangsrirat, Panyawut Sri-iesaranusorn, Warisara Asawaponwiput, Wassapon Watanakeesuntorn, Polathep Vichitkunakorn
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE43 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC13550162

Browse the full library · About the Tower of Babel