Plant diseases in the field of agriculture can cause significant loss to the farmer. This leads to decline in the quality and quantity of the crop. Hence, it is very important to identify and recognize the type of plant disease in order to help the farmer. This information can help the farmer to take appropriate decision about increase in crop not compromising the quality. Manual methods which are currently being used to detect plant diseases are said to be time consuming, since it requires expert advice and requires enormous manual effort. In order to overcome these problems new technologies are being developed, which uses computer vision and image processing techniques to detect various diseases in plant. Results have shown that these methods can produce fast, accurate disease detection and have good economic importance. This proposed work presents automatic system for classification of three important plant diseases, namely Bacterial Blight, Leaf spot and Leaf Rust. This system uses K-means Clustering for segmentation and Support Vector Machine (SVM) classifier for classification.