Classification of Skin Diseases using Digital Image Processing with MobileNetV2 Architecture

Tarisafitri, Nahla
Aljabar, Andi

Nusantara Journal of Artificial Intelligence and Information Systems
Vol. 1 No. 1 (2025): June

Skin diseases are prevalent in tropical countries like Indonesia, where geographical and climatic conditions facilitate their spread. This research aims to classify skin diseases using digital image processing with the MobileNetV2 architecture. The DermNet dataset is used to develop and test the model. Various image preprocessing techniques, including resizing, augmentation, and normalization, were applied to the dataset, which consists of 300 images categorized into dermatitis, psoriasis, and scabies. The model achieved a training accuracy of 90% and a validation accuracy of 70%, with notable success in classifying psoriasis. The findings suggest that MobileNetV2, when combined with CNN, is a promising tool for diagnosing skin diseases early and efficiently.