Early leaf diseases prediction in Paddy crop using Deep learning model
Abhishek Bajpai, Naveen Kumar Tiwari, Ashutosh Tripathi, Vibha Tripathi, Devesh Katiyar
Abstract
At present, more than 50 % of the world’s population is dependent on rice for its survival. But there are various diseases that decrease the productivity of the paddy crop. The most affecting paddy leaf diseases are Brown spot, Hispa, & Rice blast. These illnesses restrict rice plants from growing and producing as they should, which might result in significant economic and ecological losses. The harm to the crops and the losses to the farmers can both be significantly reduced if these diseases are quickly and accurately recognized at an early stage. Multiple methods have been proposed to solve this problem using different machine learning and deep Learning techniques. In this paper, we have considered four classes for the classification of the leaf category. We used deep learning techniques to detect the actual disease of affected plants. We implemented three architectures i,e. VGGNet16, RenNet101,& AlexNet. Out of these three, Alexnet has the highest Accuracy. The AlexNet model has achieved training & testing accuracy of 92.35% and 85.27% respectively in our dataset.