Detection and Classification of Rice Leaf Diseases Using Image Processing
Keywords: Deep Learning, Image Segmentation.
Diseases of the rice leaf are a significant cause of financial and production losses in the global agriculture sector. In this study, a convolutional neural network-based image processing method is suggested for recognising illnesses of the passion rice leaf. The current packages from the front end employed in this project are used to extract rice leaf picture details in accordance with the CNN technique. However, it can take a while. Therefore, the proposed approach can be utilised to swiftly and automatically detect illnesses in rice leaves.
The essential steps of this suggested method are as follows: obtaining the input image, Preprocessing of images, Locating the afflicted areas, highlighting them, checking the training set, and displaying the results. only few different rice leaf diseases. The algorithm was used to find the rice leaf disease. Instructive images were made available. Images were turned into color models prior to image processing in order to determine the best color model for this strategy. The model was created using the Support erosion approach and the Local Binary Pattern for feature extraction. This method can identify rice leaf diseases with an average accuracy of 79% and their stage with an average accuracy of 66%.