Disease Detection in Plants using a Pseudo Color Co-Occurrence Matrix

This study reports a color based texture classification for a machine vision system for the identification of disease in plants from color leaf images. We applied the texture features in literature and studied which subset will be effective for Mango and Tomato plants. Effectiveness of each statistical functions were studied in classifying the pattern using a Support Vector Machine. For textures which are different like smooth new leaves, dry leaves and growth Gray Level Co-occurrence based statistics was effective but values failed to discriminate in tomato diseases. We propose a novel method which uses second order statistics on a pseudo color based co-occurrence matrix which resulted in a better classification for three tomato diseases. This method can be applied for early Disease Detection for any plant and help farmers take corrective measures to avoid loss of yield.