Abstract:
Segmentation of the different human skin is possible by a steadfast human skin detection method that is flexible to different human skin colors and elucidation conditions. Many human skin color detection techniques are readily avail but they are prone to error when it comes to the matter of variety of human skin colors across different racial. The existing method are difficult to analyze because of their high computational cost . In this paper, a novel human skin detection approach is used for combining a smoothed 2-D histogram with Gaussian model, for automatic human skin detection in color image(s). The skin model for the specific person is performed using an eye detector. In the proposed approach, computational cost is reduced as no training dataset is required, and it improves the accuracy of skin detection despite wide variation in civilization and elucidation. Qualitative and quantitative results on three standard public datasets and a comparison with state-of-the-art methods have shown the efficiency and heftiness of the proposed approach.

Keywords:Skin detection, Color space, dynamic threshold, fusion strategy