Robust human face detection in complicated color images

• Published in: 2010 2nd IEEE International Conference on Information Management and Engineering pp. 218 - 221
• Main Authors: Jiang Qiang-rong, Li Hua-lan
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.04.2010
• Subjects: Color; Eyes; Face detection; Humans; Image edge detection; Image segmentation; morphological operation; Robustness; Skin; skin color segmentation; Surveillance; Video sequences; YCbCr color spaceskin color model
• ISBN: 9781424452637; 1424452635
• DOI: 10.1109/ICIME.2010.5477567

In recent years, the research on detecting human faces in color image and in video sequence has been attracted with more and more people, but automatic human face detection from images in surveillance and biométrie applications is still a challenging task due to the computation inaccuracies and the continuous nature of some transformations. In this paper we propose a novel face detection algorithms based on combining skin color model, edge information and features of human eyes in color image. First, the elliptical skin model in YCbCr color space is constructed to segment skin color pixels from the background image in conjunction with edge information. Then we can use mathematical morphological operators to fill holes in the regions, and extract candidate face regions. Finally, face verification is employed to determine whether candidate regions are face regions. Experimental results show the accuracy and robustness of the algorithm by testing a great deal of face color images.