RICE LEAF DISEASE CLASSIFICATION USING CNN WITH TRANSFER LEARNING

Rice is one of India’s largest commonly cultivated crops, and it is vulnerable to a variety of illnesses at different stages of production. With their inadequate understanding, farmers find it extremely difficult to manually diagnose these illnesses. Recent advances in Deep Learning have shown that automatic image recognition systems based on convolutional neural network (CNN) models can be quite useful in these cases. Because a collection of rice leaf disease images is not readily available, we constructed our own tiny dataset and developed our deep learning model using Transfer Learning. The suggested CNN architecture is based on VGG16, and it was trained and tested using data from rice fields and the internet.