Rice Leaf Diseases Classification Based on RiceNet

This paper presents a new model called RiceNet, which is applied to classify rice leaf diseases with compatible and improved accuracy. The RiceNet model is built with 30 layers that are responsible for collecting and extracting features from input diseased leaf images. The research experiment was performed on the Paddy Doctor dataset, which includes 10,407 rice leaf disease images labeled with nine disease types and one standard, respectively. The model’s performance was evaluated on the correct type of information classification and implemented for various cases. Research results show that the RiceNet classification model has an accurate classification rate of 93.08%.