Abstract: In order to segment the information of skin color in the image effectively, we proposed a algorithm of skin color segmentation based on texture and neighborhood information of skin point. Firstly, according to the distribution of skin color point in the image in RGB color space, we used the statistical histogram model to segment the part of skin color in the image initially. And then, in order to extract the texture information in the image, we used the Gabor filter to further filter out the part of closing to human skin color but the texture was rough. Finally, we used the  skin color point in the mask image as the seed point to diffuse neighborhood information, and then got the final mask image. The experimental results show that, compared with the skin color model of statistical histogram, the method is better in skin color segmentation.

Key words: Gabor function, skin color segmentation, statistical histogram model, RGB color space, texture information