Volume 2 - Issue 3, May - June 2026

The occurrence of plant diseases has become an integral part of influencing agricultural crop yields. Detecting plant diseases helps minimize the losses that occur in the agricultural industry. Detecting plant diseases is a time-consuming activity that requires the involvement of several agricultural experts. It is not feasible to do this in rural areas owing to the scarcity of such experts. This research discusses the application of an efficient approach to identify plant diseases using lightweight convolutional neural networks (CNN). The details of an effective model to diagnose plant diseases along with their remediation are provided here. The architecture of the neural network utilizes inception blocks, residual networks, and depthwise separable convolutions that help reduce the computational burden. These deep learning models can be trained using freely available image data sets of leaves that are either healthy or diseased. The proposed CNN-based model performs better than other deep learning models at reduced computational costs. The design of the application software interface is used to identify and mitigate plant diseases in real-time.

Kawsheka MJ,"Convolutional Neural Network-Based Approach for Plant Disease Identification" International Journal of Advanced Multidisciplinary Research and Educational Development, V2(3): Page(321-323) May-June 2026. ISSN: 3107-6513. www.ijamred.com. Published by Scientific and Academic Research Publishing.