<![CDATA[This study examines the application of Convolutional Neural Network (CNN) algorithms in detecting pests on water spinach plants using two CNN architectures: MobileNetV2 and VGG16. Monitoring diseased leaves during the plant growth phase is a critical step, and AI-based solutions are essential to enhance the accuracy of automated pest detection. In this research, a dataset of water spinach leaf images, both pest-infected and healthy, was collected to train the two CNN architectures. MobileNetV2 was selected for its ability to deliver high performance with low computational complexity, while VGG16 was used as a benchmark due to its deeper architecture and widespread use in various image recognition tasks. The testing results indicate that MobileNetV2 achieved a detection accuracy of 84%, while VGG16 yielded an accuracy of 83%. Thus, MobileNetV2 is considered superior for this pest detection application as it provides a balance between high accuracy and optimal computational efficiency. The study concludes that MobileNetV2 is a more suitable architecture for pest detection systems in water spinach plants, particularly for applications requiring high performance on resource-constrained devices.]]>
Copyrights © 2025
