摘要
人脸识别是计算机视觉和模式识别领域的一个活跃课题,有着十分广泛的应用前景.主要研究了人脸识别技术中的人脸特征提取和特征比对技术.在特征分类中,距离测量选择了最佳距离度量最近邻法,提出了一种新颖的测量方法——优中选优,可以使分类器的设计更加简洁、有效,使用较少的特征向量数目就能取得较高的识别率.仿真结果表明了该方法的有效性.
The face recognition is an active subject in the fields of computer vision and pattern recognition, having a wide range of potential applications. This thesis focuses on the problems of face feature extraction and face compare in face recognition technology. In the feature classification mentioned, the method of the shortest distance, and the nearest neighbor rule are used in distance measurement. The algorithm proposes a novel method of measurement-optimization-selecting the best from the good ones, the Classifier in this algorithm can be simplified to make it more compact and effective, and higher correct recognition rate can be gained using less number of feature vectors. The effectiveness of the approach is experimentally demonstrated.
出处
《应用科技》
CAS
2009年第2期42-45,共4页
Applied Science and Technology
