Fruit diseases cause destruction in agricultural field and cursing economic loss. Identifying the fruit diseases in manual is not efficient and it produces faulty results. A solution to this problem is suggested. The proposed system uses both image processing and data mining techniques. The image processing technique is used to identify the discrepancies in the fruits based on their color and texture property. These discrepancies can be rectified by data mining technique like feature extraction and classification method. The features are extracted from the image using Color coherent Vector, Local Binary Patterns algorithm. The extracted features are passed into classifier. This proposed system using K-nearest neighbor (KNN) algorithm for classification, improves the statistical estimation of fruits with disease and pattern recognition. The experimental results show the proposed system is supporting efficiently the accurate detection and identification of fruit diseases.

computer vision, statistical analysis, image segmentation, local binary pattern, color coherent vector, classification.