人脸检测与识别

人脸是人主要的身份特征之一,人脸识别在信息管理、身份验证、视频监控、机器视觉等领域有着重要的实用价值.本文以中央民族大学信息工程学院本科生人脸数据库为背景,从应用的角度出发,开发了基于人脸识别的学生身份确认系统.整个系统主要分为图像预处理、人脸检测和人脸识别三个部分.图像预处理包括图像大小归一和图像灰度均衡;人脸检测采用Harr分类器;利用PCA（主成分分析）获得投影矩阵和维数适中的能够表征人脸的特征向量;基于最近邻原则,进行人脸匹配.系统测试结果表明,该系统能够完成基于人脸识别的学生身份信息查询功能.

Face Detection and Recognition

Human face is one of the main identity characteristics, and facial recognition has important practical value in the field of information management, authentication, video surveillance, machine vision. From the application point of view, this paper, aiming at the face database of the undergraduate students from the School of Information Engineering, MinZu University of China, has developed the student identity verification system based on face recognition. The system consists of three parts including the image preprocessing, facial detection and facial recognition. The image preprocessing includes image size normalization and image gray balance. In the facial detection, Harr Classifier is applied. By use of PCA（ Principal component analysis） , the projection matrix from the original facial space to the facial feature space is achieved and the facial feature vector with appropriate dimension is obtained. Facial matching is carried out based on the principle of the nearest neighbor. The result of the system testing demonstrates that the system can completely fulfill student＇s information query based on the facial recognition.