【作者】 汪荣贵； 孙见青； 胡韦伟； 李守毅；

【Author】 WANG Rong-gui SUN Jian-qing HU Wei-wei LI Shou-yi Faculty of Computer and Information, Hefei University of Technology, Hefei ,230009, China

【机构】 合肥工业大学计算机与信息学院；

【摘要】 为提高人脸检测速度,提出了一种基于特征和基于图像相结合的快速人脸检测方法。在训练阶段,利用训练样本的低频小波系数来训练fisher线性分类器和支持向量机(SVM);在检测阶段,首先利用双眼区域的亮度关系和脸部的对称性来快速过滤掉大量的背景区域,再利用fisher线性分类器去除部分背景区域,最后用SVM对余下的区域进行进一步的验证,以确认是否是人脸。实现结果表明了该方法的正确性和有效性。更多还原

【Abstract】 In order to improve the speed of detecting human face, an algorithm of fast face detection combining both feature-based and image-based is presented. In the training step, the fisher linear discrimination and SVM are trained using the feature extracted by Wavelet transform. In the detection step, firstly, the characteristics of intensity in eye region and symmetry in face are used for finding face candidates, then use the fisher linear discrimination to exclude large parts of backgrounds, and use the SVM to perform the final detection. Experimental results show that the new algorithm is feasible and efficient.更多还原