Abstract: Quantitative evaluation of skin texture or micro-contour is of great significance for evaluating the efficacy of cosmetics against wrinkles.Skin texture,as one of the inherent characteristics of the skin,has a significant effect on skin quality.Based on the application requirements of skin beauty and the actual needs of daily life,the basic texture features of the skin were studied by the related algorithms in the field of image processing.Firstly,the experimentally measured skin image was transformed into a grayscale image.The image was then enhanced by contrast limited adaptive histogram equalization.Then the noise of the image was removed by Gauss filtering,and the details of the texture were enhanced by Wiener filtering.A clear texture of the skin image was obtained.The gray scale and distance of gray level co-occurrence matrix suitable for skin texture evaluation were determined through experiments.The skin texture was statistically analyzed based on gray level co-occurrence matrix algorithm.A mathematical model of comprehensive index based on four texture feature parameters was proposed,and the texture features of all skin images were quantitatively evaluated by using the model.Visual evaluation of these skin images was also conducted by experts.The two methods of evaluation are in good agreement.

Key words: cosmetic efficacy evaluation, skin texture, image preprocessing, gray level co-occurrence matrix algorithm, synthesis index