Skin feature extraction and processing model for statistical skin age estimation

The convergence of information and medical technologies has resulted in the emergence and active development of the ubiquitous healthcare (U-Healthcare) industry. The U-healthcare industry provides telepathology and anytime-anywhere wellness services. The main purpose of these wellness services is to provide health information to improve the quality of life. Human skin is an organ that can be easily examined without expensive devices. In addition, there has recently been rapidly increasing interest in skin care products, resulting in a concomitant increase in their consumption. In this paper, we propose a new scheme for a self-diagnostic application that can estimate the actual age of the skin on the basis of the features on a skin image. In accordance with dermatologists’ suggestions, we examined the length, width, depth, and other cell features of skin wrinkles to evaluate skin age. Using our highly developed image processing method, we could glean detailed information from the surface of the skin. Our scheme uses the extracted information as features to train a support vector machine (SVM) and evaluates the age of a subject’s skin. Evaluation of our proposed scheme showed that it was more than 90% accurate in the analysis of the skin age of three different parts of the body: the face, neck, and hands. Therefore, we believe our model can be used as a standard or as a scale to measure the degree of damage or the aging process of the skin. This scheme is implemented into our Self-Diagnostic Total Skin Care system, and the information obtained from this system can be utilized in various areas of medicine.