A New Robust Face Detection Method Based on Corner Points
Samir El Kaddouhi, Abderrahim Saaidi, Mustapha Abarkan · International Journal of Software Engineering and Its Applications · 2014
This paper describes a method of faces detection based on skin color, face shape and corner points. The skin regions are detected by the segmentation of image with a threshold calculated by a combination of rules proposed on elements of the three color spaces RGB, HSV and YCbCr. Next, the geometric constraints (surface, ratio, eccentricity) are applied to eliminate the skin regions detected that do not have a shape similar to a human face. Then the corner points of Harris are used to verify the existence of eyes in the regions detected. The results obtained are satisfactory in terms of the quality of detection and rapidity.