Skin Color Detection

614 Words Feb 24th, 2018 2 Pages
The idea behind using this technique is that skin color is different from other object’s color and then can be used to identify human gesture part. There are several color spaces such as RGB, YIQ, YUV, and generic color model that are used to distinguish skin pixels from non-skin pixels through comparing a predefined skin color model (skin pixels) with pixels of the input image. Figure 2 shows skin color technique using YIQ and YUV color spaces.
Thermal camera is an infrared camera that is able to detect infrared radiation emitted from the human body (temperature). It can determine the human gesture region depending on the human temperature while background is neglected. Figure 3 shows human interaction captured by thermal camera.
This technique is used to track and detect moving gesture parts depending on the difference between two or three consecutive video frames. The pervious image (frame) is subtracted from the current image (frame) in order to obtain the moving part.
A. Background subtraction
The idea behind this technique is to set a background as a reference or model and then compare the current frame with the reference background in a pixel-by-pixel manner to detect the moving foreground object. The background model is updated with a new image from time to time in order to track dynamic changes. Figure 4 shows background subtraction…