A Smart Application as Solution for Diagnosis of Rice Diseases in Pakistan: An Image Processing Approach

Abstract

Rice production in Pakistan, once a dominant force in Asia's rice industry, has seen a significant decline due to various diseases and the continued reliance on traditional farming methods. The lack of education and awareness among farmers about modern agricultural techniques and applications further exacerbates the challenges faced by the industry. To address these issues, this research presents an innovative application based on image processing techniques, offering a platform to assist farmers with information on rice diseases and disease diagnosis. The solution utilizes Convolutional Neural Networks (CNN), powered by deep learning, to enhance the accuracy of rice disease classification using a dataset from Sialkot, Pakistan. The user-friendly and reliable nature of the application eliminates the need for additional machinery or installations, making it a practical and accessible tool for the farming community. The results demonstrate the potential of the application in empowering farmers with knowledge and aiding in the betterment of crops and overall production, the accuracy of CNN is 99.9%, and fostering a path towards revitalizing Pakistan's rice industry.