Non-Destructive Mangosteen Volume Estimation via Multi-View Instance Segmentation and Hybrid Geometric Modeling
Wattanapong Kurdthongmee, Arsanchai Sukkuea, Md Eshrat E Alahi, Qi Zeng · 2025
Read Non-Destructive Mangosteen Volume Estimation via Multi-View Instance Segmentation and Hybrid Geometric Modeling by Wattanapong Kurdthongmee, Arsanchai Sukkuea, Md Eshrat E Alahi, Qi Zeng (2025) 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-12842316.txt.
Description
Wattanapong Kurdthongmee and three co-authors wrote this 2025 study on sizing mangosteen fruit without cutting it open. Growers grade fruit by volume, but the mangosteen's big leafy cap breaks the simple shape formulas used for rounder fruit. The team photographed fruit from several angles and taught a vision model to outline the fruit body apart from its cap.
Leaving the cap in the outline wrecked the volume estimates, so the separation step proved vital. They then trained eight math models on hundreds of images, mixing side and bottom areas with a standard oval-shape estimate. Models using several views beat single-view ones, and the best hybrid cut the average error to about sixteen percent. The setup is cheap and fast enough for sorting lines. It offers small farms a grading tool built from plain cameras.
Catalog record
| CREATOR | Wattanapong Kurdthongmee, Arsanchai Sukkuea, Md Eshrat E Alahi, Qi Zeng |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |