Penerapan Model Transfer Learning Dalam Mendalami Penyakit Daun Jagung Menggunakan Arsitektur VGG19

Setiyo Adi Wibowo
Rudi Kurniawan
Budi Santoso

Journal of Informatics, Electrical and Electronics Engineering
Vol. 5 No. 1 (2025): September 2025

The process of processing corn leaf disease data using the VGG 19 architecture based on deep learning is to analyze corn leaf diseases that result in low yields. In describing the values to be managed in this study, a digital image dataset of corn leaf diseases consisting of 5 classes with 3923 images per class was used. The objectives of this study are to enable easy prediction of corn leaf disease and to treat the disease. It also aims to enable pattern recognition of corn leaf disease based on digital images using the VGG19 architecture model. The results of corn leaf disease classification obtained from the VGG19-based model show excellent performance in identifying various plant health conditions. With an overall accuracy of 97.96%, this model successfully distinguishes between five disease classes, namely Common Rust, Grey Leaf Spot, Healthy, Northern Leaf Blight, and Northern Leaf Spot. This figure reflects the effectiveness of the model in recognizing the distinctive visual patterns of each disease, which is very important for effective crop management.