摘要
针对智能会议场景对人脸识别的特殊情况,通过依据检测、跟踪得到头部区域与人脸区域的面积比,选择正面的人脸进行识别,降低了人脸姿态对人脸识别的影响.在分类方法的选择上,采用支持向量机方法,并对支持向量机方法进行了概率建模,分类器输出结果是测试人脸属于每类的概率.实验结果表明:该方法不仅使人脸识别的精度得到了提高,还提供了其属于所在类中的可信程度.
To analyze face recognition in intelligent surveillance, face recognition based on the probability outputs of multi-class SVMs is proposed. Considering the special situation for face recognition in intelligent meeting scene, the front face is chosen for face recognition to reduce the pose effect, which is the ratio of the lengths or areas between the head and the face. The SVM method, which is modeled by the probability, is chosen as the classification method. The output of the SVM classifier is the probability of the tested face in each class, and the recognized face denotes the class with the highest probability. Experiment results show that this method not only improves the precision of face recognition, but also provides the reliability of the classification.
出处
《武汉理工大学学报（交通科学与工程版）》
2009年第2期345-348,共4页
Journal of Wuhan University of Technology（Transportation Science & Engineering）
基金
国家自然科学基金项目(批准号:60672137)
教育部高校博士点基金(批准号:20060497015)
湖北省自然科学基金项目(批准号:2004ABA043)
中国石油科技创新基金项目(批准号:2008D-5006-03-04)资助
关键词
智能监控
概率
支持向量机
识别
intelligent surveillance probability support vector machines recognition
