ABSTRACT
Modern agriculture is an evolving approach to agricultural practices and innovations that help farmers to reap maximum yield by systematically managing resources and diseases. Proactive disease management helps to predict and diagnose diseases in advance, to take appropriate measures and effectively treat the plants. In our paper, we have proposed a proactive plant disease prediction system which uses deep learning models Convolution Neural Networks. Using CNN makes the system more robust, unlike variations observed using machine learning algorithms. We have achieved a prediction accuracy of average 93%, considering seven different plant diseases occurring in this region. By, using microscopic images, we can use classification problem for prediction. The pathogenic activity which starts much before the disease fully develops can be analyzed using microscopic images, which we have exploited in predicting disease.
