Minimizing false detection of skin color by using background subtraction

Detecting skin color from image sequences has played an important role in many computer vision applications, such as face detection and tracking. Even though there are a number of approaches for skin color detection, most of the existing algorithms produce false positives when non-skin pixels have similar color to skin color. To overcome this problem, this paper proposes a method that combines skin color detection with background subtraction. In experiment, we quantitatively evaluated three existing skin color detection methods with mixture of Gaussian-based background subtraction. Experimental results show that the proposed approach significantly reduces false positives.