Rice Crop Disease Forecasting Using Tuned Deep Learning Approach

Plant disease detection plays an important role in preventing losses due to these diseases. The majority of plant infections occur as leaf spots and can be detected from leaf images. In this paper, A model based on convolutional neural networks is suggested for identifying diseases in rice crops. To assess the performance of the models, accuracy, dropout, and loss function are utilized. The implemented models achieved an accuracy of 99.12, and 99.73 for InceptionResNetV2 and ImpShuffleNet(Proposed Model) respectively. It is also observed from the results that the training time and average time per epoch have been reduced for these two models.