Explainable Leaf Disease Classification

Lightweight hybrid-feature pipeline for plant disease detection

A lightweight and explainable pipeline for plant leaf disease classification, built on handcrafted hybrid features rather than a large end-to-end network — designed for deployment on low-resource hardware in agricultural settings.

The design prioritizes interpretability and computational efficiency: every feature has an agronomic meaning, and the model is small enough to run without a GPU.

Stack: Python, scikit-learn, OpenCV, explainability tooling

View on GitHub