Diagnosis of pomegranate plant diseases using neural network

Pomegranate is a fruit which grows with a very high yield in many states of India and one of the most profits gaining fruit in the market. But due to various conditions, the plants are infected by various diseases which destroy the entire crop leaving very less product yield. So, the work proposes an image processing and neural network methods to deal with the main issues of phytopathology i.e. disease detection and classification. The Pomegranate fruit as well as the leaves are affected by various diseases caused by fungus, bacteria and the climatic conditions. These diseases are like Bacterial Blight, Fruit Spot, Fruit rot and Leaf spot. The system uses some images for training, some for testing purpose and so on. The color images are pre-processed and undergo k-means clustering segmentation. The texture features are extracted using GLCM method, and given to the artificial neural network. The overall accuracy of this method is 90%. The results are proved to be accurate and satisfactory in contrast to manual grading and hopefully take a strong rise in establishing itself in the market as one of the most efficient process.