In plants, disease detection is a significant task related to the agricultural production of a country. The major economy of a country is linked with agricultural production which is very important for the development of a nation. Any kind of infection on the leaves of the plants leads to a loss in crop production and puts down the effort of farmers which in turn hits the economy and livelihood of the country. In this paper, we propose an image processing and filter-based feature selection method which distinguishes the disease of the plant leaf and classifies them using a KNN classifier. Using the feature selection process the unwanted redundancy and irrelevant data is filtered which helps the classifier to learn and classify the data more precisely.