Segmentation of faces in video footage using HSV color for face detection and image retrieval

Previous studies on face detection in video footages show that segmenting faces accurately and reliably is often hard to succeed, leading instead to laborious and tedious interactive manipulation. This paper presents a segmentation method using controlled weights on the three HSV components and constructs a face detection and image retrieval system. First, it is shown that HSV color has advantages over RGB or YCbCr one when segmenting a face and generating a binary pattern that retains as many features of the face as possible in the original color picture. Then, a face detection and image retrieval system is constructed using HSV color, where each time a significant scene change is detected segmentation is carried out for the beginning frame using a few sets of the weights on the HSV components, and resulting patterns are checked in some requirements and correlated with a typical face pattern. Computer experiments show that the successful detection rate is more than 95 percent and that images can be retrieved from an input face image in a short time.