Agriculture is an essential source of livelihood in the entire world. It generally contributes 17–18% of India’s GDP. In the early days, there were many challenges, like low productivity, land degradation, etc. Manual identification of plant disease is difficult, and so modern agricultural techniques try to improve productivity. If timely diagnosis and precise analysis are maintained for the plant, then there might be a reduction in diseases, and improvement in health can be observed with the increase in output. ML and DL models offer a variety of techniques for the identification of diseases through the use of applications. In this paper, the DenseNet121 model is employed to identify rice plant diseases. Rice plant diseases are classified into 11 different diseases. The study includes 11,000 images of rice plants. We have applied the DenseNet121 model in deep learning. For increasing the classification performance, parameters like the number of epochs and learning rates are decided. Also, the parameters of the classification report are compared for different considered classification models like VGG16, AlexNet, and DenseNet121. Proposed model accuracy is increased to 99.89% and the training time for 11,000 images is also decreased.