HUMAN SKIN DETECTION USING IMAGE FUSION

Skin detection is a process of finding skin colored pixels and regions in an image or a video. Skin detection is a primary step in variety of image processing applications right from face tracking, gesture analysis, video surveillance, image content filtering, and video annotation to various human-computer interaction domains. Effective skin detection is the process of identifying human skin colors of different ethnic and under different illumination conditions. Different skin detection however has been successfully applied but they are prone to false skin detection and also not able to cope up with the variety of human skin colors across different ethnic. Additionally requires high computational costs. In this project a dynamic threshold approach is proposed which would reduce computational costs by eliminating training stage to improve the accuracy of skin detection despite wide variation in ethnicity and illumination. An eye detector would be used to refine the skin model for a specific person. Additional feature would be implemented i.e. fusion of a smoothed 2-D histogram and Gaussian model for automatic human skin detection in color image(s) to improve the results of skin detection.