An Automated Rice Leaf Disease Identification using Deep Learning

Extensive research and modern advancements have proven that Artificial Intelligence (AI) has brought a drastic change for all the sectors of the economy including the rural agricultural sector. This research paper is focused at applying the method of Deep Learning (DL) for rice leaf disease identification - a major problem in the farming industry. This research project incorporates methods like data preprocessing, model training, and evaluation with the help of deep learning architectures like Tensorflow, transfer learning, Conv2D and BatchNormalization. Thus, with a combination of these intelligent techniques, the contribution of this research paper is that the proposed custom model has acquired an accuracy of 95% while using noticeably lesser number of layers when compared to some of the popular pre-trained AI models like MobileNet, ResNet and InceptionV3.Deep Learning (DL) has immense potential to demonstrate the benefits of AI in farming and agriculture. There are countless opportunities to detect rice leaf diseases early with the objective to improve crop management and productivity through Artificial Intelligence (AI). Keeping this in mind, the data for training AI model has been optimized using ImageDataGenerator followed by strategic callbacks. The result speaks for itself as it provides precise accuracy to identify rice leaf disease. Additionally, this research paper talks about precision agriculture and its importance in the modern agricultural industry. It highlights the prospective to increase farming efficiency, reduce resource wastage, and foster sustainability in agriculture. Through this research, it is evident that the utilization of AI in the agricultural sector can facilitate informed decision-making, resource optimization, and the early detection of plant diseases.