- Home
- Archives
- 2026
- Vol. 6 Issue 6
- Article
AI Powered Smart Fruit Quality Analysis and Ripeness Prediction System with Horticultural Advisory
Authors
1 2 3 4 Department of Information Science and Engineering, Bapuji Institute of Engineering and Technology, Davanagere, Karnataka, India. 5 Assistant Professor, Department of Information Science and Engineering, Bapuji Institute of Engineering and Technology, Davanagere, Karnataka, India.
Article Information
- Published In
- Volume 6, Issue 6 (2026)
- Published Online
- November-December 2026
- Pages
- 211-215
- Article Type
- Original Article
Abstract
View PDFThe developed system is a browser-based fruit analysis prototype that combines image processing, a compact TensorFlow.js classifier, and an optional server-mediated generative analysis path. A user selects an image or captures one with a camera. Client-side Canvas routines derive HSV colour ratios, CIELAB descriptors, and shape measures, while a MobileNetV2 feature extractor and locally supplied dense head select among eight fruit classes. Ripeness, consumption status, and several advisory values are produced through fixed rules and metadata rather than a separately validated maturity model. When configured, an Express endpoint can send an image to Gemini and return fields that may supplement or override the local result; on failure, the interface retains a local result. The application also includes account screens, a chat interface, local scan history, and model-calibration and training demonstrations. Inspection of the supplied files found a 96-image sample collection and a validation report that reuses source images included in training, so its 96/96 result is not evidence of independent accuracy. This paper documents the implemented workflow and its limits; no independent field, ripeness, or performance evaluation is available in the archive.
Related Articles
Share Article
*Instagram doesn't support direct link sharing from web. Copy the link and share it in your Instagram story or post.