<![CDATA[Potato plants are an important food crop but are susceptible to leaf diseases such as early blight and late blight, which can significantly reduce crop yields. In this study, we developed and compared several convolutional neural network (CNN) models to classify potato leaf diseases based on visual images. The data used consisted of potato leaf images in three classes: healthy, early blight, and late blight. The image dataset was processed through augmentation and normalization to improve model accuracy. Three CNN architectures, namely MobileNet-V2, VGG16, and ConvNeXtBase, were implemented and tested with different batch sizes. Based on the results, the VGG16 architecture with a batch size of 32 provided the best performance with a classification accuracy of 95.93%, followed by MobileNet-V2 with an accuracy of 94.15%. Therefore, CNN models, particularly VGG16, proved effective in identifying potato leaf diseases, contributing to more efficient crop management and reducing yield losses.]]>
Copyrights © 2024
