The main objective of our study was aimed at quantifying the severity of disease symptoms Cercospora, Anthracnose and Alternaria alternata in Dalbergia sisso, Luffa actangula and Solanum melongena leaves. The first step of the proposed algorithm was the capture of the diseased region using digital camera followed by pre-processing stage to remove the unwanted background, noise and poor resolution of image. K- mean clustering algorithm was then applied to divide the image into respective clusters followed by extraction of number of color and texture features using Gray Level Co-occurrence Matrices (GLCMs). These parameters were finally fed to Back Propagation Neural Network (BPNN), which contain a hidden layer between the input and output images which performs the final classification.
