MUNGBEAN LEAF DISEASE DETECTION USING K-NEAREST NEIGHBOR ALGORITHM

Abstract

Disease detection is a very important part to protect loss of crop in agriculture. Symptoms of the plant diseases can be detected by using machine learning techniques. Machine learning technique can solve for classification and regression problems. This proposed system presented that mungbean leaf disease detection by using digital image processing and machine learning techniques. Image preprocessing state used image enhancement technique to improve the quality of images. This enhanced image is segmented by using k-means clustering techniques. This technique is used to segment region of interest in leaf area. Gray Level Co-occurrence Matrix (GLCM) is used to extract features from preprocessing and cluster images. And also, mungbean leaf diseases are classified using the k-nearest neighbor algorithm (k-NN). According to the results of the experiments, the system can successfully detect and classify healthy and unhealthy or infected leaf areas. In this system, the k-NN algorithm can classify disease types with 96.7% accuracy and the support vector machine (SVM) algorithm with 86.7%.