Face Detection Based on Fractal and Complexion Model in the Complex Background

摘要

In this paper, a new face detection algorithm based on fractal theory and complexion model is proposed. On the basis of the traditional face detection, the new algorithm estimated the fractal dimension of face area to exclude the non-face area and extract the face area fast and accurately. Furthermore, its calculation is simple and it can realize multi-angle detection in the complex background. As well as the new algorithm can overcome the shortcomings of high misdetection and low capability in adapting to changes of traditional way. Simulations are carried out with detailed analyses, which demonstrating the superior accuracy and high efficiency of our algorithm.