<![CDATA[Cassava plants play an important role as a national food source. However, their productivity has declined in recent years due to leaf disease. Manual disease identification is often inaccurate and slow. This study aims to develop an automatic classification system based on digital images to detect cassava leaf disease quickly and accurately. The method used is a Convolutional Neural Network (CNN) with a VGG16 architecture. The system was developed following the CRISP-DM approach and uses tools such as Python, Keras, TensorFlow, and TensorFlow Lite for integration into Android. The model was trained to recognize five leaf conditions: brown spots, bacterial blight, green mite, mosaic, and healthy. Testing over 50 epochs showed an accuracy of 96%, with precision, recall, and F1-score ranging from 0.93 to 0.98. This approach is superior to the research by Setyanto and Ariatmanto, which only achieved an accuracy of 72.84%. This system helps farmers perform early diagnosis by taking or uploading photos of leaves, enabling more effective disease control.]]>
Copyrights © 2025
