SMART AI-BASED FACIAL SKIN ASSESSMENT USING SR-GAN AND MASK R-CNN

Facial skin assessment is an essential part of dermatology, skincare analysis, and early diagnosis of skin abnormalities. Traditional assessment methods depend on expert evaluation and controlled imaging environments, which makes them inconsistent and less accessible. This research proposes a Smart AI-based facial skin assessment system that integrates Super-Resolution Generative Adversarial Networks (SR-GAN) for enhancing low-quality facial images and Mask R-CNN for accurate facial skin segmentation. SR-GAN reconstructs high-resolution texture details such as pores, wrinkles, acne, and pigmentation, while Mask R-CNN isolates only the relevant skin regions by removing background, hair, and noise. The combined pipeline improves diagnostic clarity, segmentation accuracy, and texture-level feature extraction, enabling reliable AI-driven skin analysis. This system is suitable for tele-dermatology, cosmetic evaluation, mobile skincare applications, and real-time skin-health monitoring.