Automatic Detection and Classification of Brinjal Leaf DiseaSES

Abstract

The diseases in the Brinjal can be identified through the symptoms occur in Brinjal leaf. The indication in touch difference bin of various plant diseases. The designation of disease detection need the specialist's opinion. The inappropriate identification can result in tremendous quantity of economic loss for farmers. Rather than manual identification, computers are accustomed to give automatic detection and classifying differing kinds of diseases. In this paper, lesion areas affected by diseases are segmented using different techniques, namely DeltaE, Otsu, FCM, k-means algorithm are employed. The proposed method is the image blend by discrete wavelet transforms to increase the excellence of image and reduce uncertainty and redundancy for identification and assessment of agricultural yield which can be done by DeltaE. Further color, texture and structural based features are mixed collectively for getting better performance when compared with single feature extraction.