Improving Cotton Leaf Disease Detection with MobileNetV2: An Analysis in Comaparison with MobileNetV3-Small

Abstract

Cotton leaf diseases need to be recognized and categorized in order to improve cotton production and guarantee high-quality yields. The increasing impact of climate change on disease transmission requires the use of accurate and effective early detection methods. This study compares the performance of two compact convolutional neural networks, MobileNetV2 and MobileNetV3Small, in identifying cotton leaf illnesses. While MobileNetV3Small shows an accuracy of 0.75, whereas MobileNetV2 attains an accuracy of 0.99. This significant distinction highlights MobileNetV2 has improved ability to discriminate between healthy and diseased leaves, making it an essential tool for practical and scalable application in accurate farming. According to these findings, MobileNetV2 can provide quick interventions, the prevention of disease, and an overall rise in cotton yield.

Topics & Concepts