India has the largest area of land under rice cultivation, however, its manufacturing per hectare is comparatively low compared to other countries. Even though the production of rice is increasing globally, pathogens and pests reduce the production and the rice quality causing substantial economic losses and a reduction of food security. Hence, accurate detection of disease is an essential part of food security and disease management. In this paper, we aim to develop a method of rice disease detection using neural network-based algorithms. The method focuses on the establishment of a deep learning model for the verification and identification of diseases in rice plants. Deep Learning which is a subset of machine learning can be implemented in crop disease diagnosis with improved accuracy. A technique for detecting diseases has been established using ANN and pre-trained CNN-based architecture namely VGG16, VGG19, InceptionV3 and MobileNetv2 for four different classes of rice disease with a comparative analysis between original datasets. Further, methods of automatic disease detection have also been incorporated using the Internet of Things platform with the use of ESP 32 cam which also implements MobileNetV2 as the neural network architecture. The work can substantially benefit the agricultural sector due to its improved accuracy of 99.14% using the pre-trained learning model VGG16.