Rubadevi GDivya R2025-03-152024-04-1023673370https://dspace.psgrkcw.com/handle/123456789/5405One of the most popular fruits in the world, bananas are particularly vulnerable to different leaf spot diseases, which causes large financial losses in the banana industry. Pest indicators have been necessary to identify illnesses early and spare farmers from financial damage. This issue will have a direct impact on the nation’s economy and the productivity of bananas overall. In this work, three common diseases of banana leaves—Pestalotiopsis, Sigatoka, and Cordana leaf spot—are diagnosed using the quick and lightweight IGO-SNet. The Improved Grasshopper Optimization (IGO) technique is used to optimize the SNet parameters. In the fundamental GOA, the IGO algorithm incorporates gravity force into each grasshopper’s updated position. Every grasshopper updates its position with velocity, and the new position is calculated by adding the velocity to the current position. Then, based on likelihood, each grasshopper accepts the modified position in the most appropriate manner. Diseases of banana fruit and stem can also be detected by the IGO-SNet model. It will enable growers to halt the spread of illness and take preventative measures to avoid losing banana yield. The results demonstrate that the suggested system has successfully diagnosed banana leaf illnesses with 96.53% accuracy, 96.25% recall, 96.17% F1-score, and 95.13% precision for banana leaf spot diseases (BananaLSD dataset).en-USOther