Leaf Disease Detection: Feature Extraction with K-means clustering and Classification with ANN

Agricultural productivity plays a major role in an Indian economy; therefore the disease detection in the field of agriculture is important. Farmers struggle a lot for proper crop production due to multiple diseases affecting the plant so there is a need to detect the disease at initial stage. One major disease in the crop is leaf spot. The purpose of the proposed system is to identify the leaf spot using image processing techniques. In this research the disease detection is done in four stages, image acquisition, image segmentation, feature extraction and classification. For image segmentation is done with K-means clustering method and features are computed from disease affected cluster. Features such as Contrast, Correlation, Energy, Homogeneity, Mean, Standard Deviation and Variance are extracted. The extracted features from disease cluster are given as classifier inputs to classify the disease. The classifier used in this paper is neural network (NN) classifier. It is observed that the accuracies for bacterial leaf spot and target spot of cotton leaf diseases as 90% and 80% respectively. For tomato leaf diseases- septoria leaf spot and leaf mold as 100%.