Multi-face Detection Based on Improved Gaussian Distribution

The appearances of human face, expression, color and so on are different, which bring difficulties to face detection. This paper proposed a novel multi-face detection algorithm mainly Based on improved Gaussian distribution model. Utilizing the skin-color model and facial features firstly, most likelihood regions of face candidates can be excluded quickly. Then, two Gaussian distribution curves are acquired by projecting the edge detection image to X-axes and Y-axes respectively. Finally, an accurate face region is found through the solution of Gaussian equation. Experimental results show that the scheme proposed in the paper can improve the multi-face detection accuracy and robustness.