Automated fruit maturity grading using deep learning with feature fusion
S. Kripa, V. Jeyalakshmi · 2026
Read Automated fruit maturity grading using deep learning with feature fusion by S. Kripa, V. Jeyalakshmi (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-13483444.txt.
Description
Kripa and Jeyalakshmi built a grading system for mangoes whose skin barely changes color as they ripen, defeating human sorters. Their 2026 study targeted a Thai variety that stays yellow from unripe to ripe. The system fuses camera images with internal chemistry readings, such as sugar content and acidity, inside one deep-learning model. A swarm-based tuner then adjusted the model's settings.
The fused system sorted mangoes into maturity stages with about 98 percent accuracy on its held-out test share. That beat the image-only and single-network baselines the authors evaluated alongside it. Three different image networks contributed visual features, while a small dedicated network encoded the chemistry values. Kripa notes the approach is non-destructive, since fruit is graded without being cut open, but the chemistry inputs still require sampling from each batch.
Catalog record
| CREATOR | S. Kripa, V. Jeyalakshmi |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 84 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |