A Skin Roughness Evaluation Method Integrating Local Structural Information of Skin Surface Using Deep Learning

In this paper, we propose a deep learning-based model to evaluate skin roughness, a quantitative measure of skin texture. We uniquely classified skin roughness, an unquantified index, and through this classification learning, constructed a new continuous index in a regressive manner. As a result, we confirmed that the model reached high accuracy in uniform texture images and captured detailed texture changes.