Disease Classification in Paddy Crop Leaves Using Deep Learning

The world's most important crop, paddy, provides nourishment for hundreds of millions of people. However, the area is frequently afflicted with various illnesses, which can significantly reduce crop yield and quality. Traditional methods of disease identification rely on visible inspection, which may not be accurate enough and can take a lot of time. Recent studies have demonstrated that machine learning techniques are quite effective at quickly and accurately identifying agricultural diseases. In this study, a machine learning method for categorization and image processing, which may be used to detect paddy crop diseases, is proposed. Taking pictures, interpreting images, and classifying disorders are three crucial additions to the suggested computer, in order to categorize fresh images into wholesome or diseased categories, the computer is trained using a library of images of healthy and sick rice plants. Using a digital camera or a phone, the photo acquisition aspect takes images of the paddy plants. Following the segmentation of pre-processed photos to remove the past and focus on the plant, the features such as shade, texture, and shape are extracted. The retrieved functions are then sent into the“ailment type thing,” which uses the convolutional neural network (CNN) machine learning technique. In order to investigate the characteristics that distinguish healthy paddy plants from unhealthy ones, a set of rules is trained on a labeled dataset of images. Once mastered, the set of criteria can accurately categorize new images into categories that represent either healthy or unhealthy states. The 89% accuracy of the CNN algorithm was excellent.