Human Skin Segmentation Technology Research And Application

Image segmentation is the process of partitioning the image into different not overlapped regions which each region has similar properties representing ideally different objects or different parts of an object in scenes. Image segmentation techniques were widely regarded and researched all the time by scientists.Skin region segmentation is also an important branch of image segmentation techniques, which would play a significant role.Skin color segmentation has many applications in tasks like detecting and tracking human faces and gestures, filtering web image contents and retrieving people in databases and Internet, even diagnosing diseases. Researches showed that the differences between skin pixels and non-skin pixels concentrate on luminance in despite of different race, age and gender people whose skins look different from each other. In the color space excluding luminance component different skin color distributions can be clustered. So, skin region segmentation is feasible.This paper models the skin color region with an incremental elliptical boundary statistical model. And then adapting this trained generic model to the real-life images which contain human skin to form a local skin model(or specific skin model). We can get an accurate skin region segmentation using this local skin model. Our approach comprises two major steps: (1) Segmenting specific image using generic model; (2) Updating the skin model incrementally with the available skin samples to segment image accurately.An example of human face detection shows the power of skin region segmentation using Elliptical Boundary Incremental Model which can be applied in variety areas.