摘要
提出一种基于小波变换和支持向量机的人脸检测算法,适用于复杂背景灰度图像的人脸检测。算法首先用双眼模板匹配方法进行粗筛选,之后对候选窗口用小波变换提取特征,将特征向量送入支持向量机进行分类检测。由于采用双眼模板进行粗筛选提高了检测速度,并且用小波变换提取特征向量,使特征向量的维数大大减少,从而有效地降低了分类器的训练难度。实验对比数据表明该方法具有较高的检测率和较低的虚警数,检测速度较高。
An efficient method of face detection based on wavelet transform incorporating with Support Vector Machine （SVM）is proposed in this paper. Firstly, a two-eye template matching is used for coarse filtering for speeding up, then follows the SVM classifier to make the final decision. In this template matching constrained subspace, a method of wavelet transform is used to extract feature for SVM training, which greatly reduces the complexity of training SVM. The experiment results show that the method not only achieves high detection rate, but also reduces the computational, complexity.
出处
《微计算机信息》
北大核心
2007年第34期237-238,共2页
Control & Automation
基金
河南省杰出青年基金(0412000400)
河南省教育厅自然科学基金(200410464004)
关键词
人脸检测
小波变换
支持向量机
face detection, wavelet transform, support vector machine
