It is a well-known fact that the quality and quantity of the rice crop is reduced due to plant disease. This paper proposes rice blast disease detection mechanism using Machine learning algorithm, to identify the disease in the early stage of the crop cultivation. The proposed method would find the blast disease and reduce the crop loss and hence increase the rice agriculture production in an effective manner. The images of the paddy field are captured and eight features are extracted to distinguish the healthy and the disease affected leaves. The proposed machine learning based classification methodology includes KNN and ANN. The performance of these two classification techniques is compared using an appropriate confusion matrix. The simulation results show that KNN based classification method provides an accuracy of 85% for the blast affected leaf images and 86% for the normal leaf images. The accuracy is improved to 99% and 100% respectively for the ANN based classification mechanisms.