AI-Assisted Classification of Rice Diseases and Insect Pests for Crop Growth Management

Abstract

Rice is one of the major cultivated crops in India which is affected by various diseases at various stages of its cultivation. Based on their limited expertise, farmers find it incredibly difficult to manually recognize these diseases with high precision. Recent breakthroughs in Deep Learning show that Automatic Image Recognition systems based on Convolutional Neural Network (CNN) and MobileNetV2 models can be quite useful in such situations. The proposed system harnesses the power of machine learning techniques to analyze images of rice plants and identify diseases and pests accurately. A comprehensive dataset of rice plant images, encompassing various disease and pest instances, is collected and used for training a deep learning model. The model leverages convolutional neural networks (CNN) to learn intricate patterns and features representative of different diseases and pests. Through a process of iterative training, the model achieves a high level of accuracy in classifying diseases and pests, enabling real-time detection and intervention. Early diagnosis and effective treatment of rice leaf infections are essential to maintaining the healthy development of rice plants for the attainment of a sufficient food supply that will support the increasing population. Thus, automated disease diagnostic systems can be employed to overcome the shortcomings associated with traditional leaf disease diagnosis methods, which are usually time-consuming, less accurate, and costly. Computer-aided rice leaf disease diagnosis systems are becoming more common in recent times. This paper presents various solutions based on various deep learning approaches. Based on the crop's image data and moreover it presents DL models like CNN and MOBILENET techniques with their performance measure.