Abstract:
The global population is continuously increasing, and ensuring food security for everyone is a major challenge. Rice, as one of the most important crops, plays a vital role in feeding more than half of the world's population. Early detection of diseases is crucial for effective rice cultivation. However, visually diagnosing diseases can be challenging, and by the time productivity is affected, it may be too late. This project aims to improve rice production by over 20% through early diagnosis of Rice Leaf Disease. A Convolutional Neural Network (CNN) and deep learning techniques are presented in this article to identify and categorize diseases such as Stem Borer, Sheath Blight, Rot Brown Spot, and False Smut. Leaf disease diagnosis poses a significant challenge as it can affect leaves of varying sizes. A dataset of 1045 photos was collected to train the KNN model, initially distinguishing between healthy and diseased leaves. The disease classification was then performed using CNN in the second phase. With this approach, we achieved 95% accuracy in identifying healthy leaves and 90% accuracy in detecting Sheath Blight (the highest among all diseases).
At Smart AI Technologies, we provide the best final year engineering projects with comprehensive support. As the best IEEE project center, maker, and seller in Bangalore, we offer complete working code implementation on clients' machines, structured project classes, documentation support, and assistance in research at no extra cost. Choose us for the best IEEE machine learning project experience in Bangalore.