Implementasi Metode YOLO V8 untuk Deteksi Jenis Flek pada Kulit Wajah Secara Real-Time

Abstract

Hyperpigmentation or spots are skin disorders that are quite common among many individuals. There are three common types of spots: melasma, freckles, and acne scar spots. These three types of spots often share similar visual characteristics, making them difficult to distinguish by the general public without adequate dermatological knowledge. Therefore, this study aims to develop an image processing system to assist non-experts in conducting an initial diagnosis of facial spot types, while also providing basic information on prevention and management tailored to the specific type of spot. This system utilizes the You Only Look Once (YOLO) version 8. YOLOv8 is an object detection algorithm that has seen significant improvements in terms of the balance between speed and accuracy in identifying objects in images in real-time. This research used 1,200 images, consisting of 900 for training data, 180 for validation data, and 120 for testing data. As a result, the system was able to detect three types of facial spots with an accuracy rate of 95.83%. The results of the evaluation metrics show that the model performs well in detecting types of diseases.