Hue based Plant Leaves Disease Detection and Classification Using Machine Learning Approach

Automatic segmentation and detection of leaf diseases from plant images has recently become a significant research field in world. The proposed method automatically segment and classify different leaf disease regions from the plant images. The proposed method works on three steps: In the first step, preprocessing is applied to reduce the amount of background noise in the plant image by using the Wiener filter. In the next step, The K-means algorithm is applied to segment the leaf images into uniform regions. These segments are then converted into HSV colour space and the segments whose hue value is largest are extracted as a disease segment. Finally, features are extracted from disease segment and then classified into various diseases by using SVM classifier. The proposed technique is implemented in Matlab 14a and experiments are conducted on different plant images. Our proposed system has an accuracy of 90.6%.