基于主元分析的人脸特征提取MATLAB实现
Implementation of Facial Feature Extraction based on Principle Component Analysis by MATLAB
雷松泽
人脸的特征提取是人脸识别的关键技术，采用主元分析法进行特征提取是经典的方法之一，利用Matlab进行人脸的特征提取能显著地提高计算效率。论述了利用主元分析和奇异值分解进行人脸特征提取的方法，并详细阐述其在Matlab中的实现过程，包括读取图像文件、计算均值脸、求特征值和特征向量，计算人脸特征参数。实现过程均给出了Matlab代码。实践证明利用Matlab进行主元分析提取特征是一种有效的方法。[著者文摘]
Facial feature extraction is a key technique of face recognition. Feature extraction using principle component analysis is one of the classical ways. Facial feature extraction by Matlab can increase computational efficiency. The facial feature extraction based on principle component analysis and singular value decomposition is discussed in this paper. The implementation methods by Matlab, which include reading image files, computing mean face, computing eigenvalue and eigenvector, getting facial feature, also are discussed in detail. For all this Matlab source codes are presented. Practice shows that feature extraction using PCA by Matlab is an effective method. And this method is useful for the researcher in this field.[著者文摘]
人脸特征识别
Human face recognition
郭晶磊 韩其睿
人脸识别是模式识别，图像处理等学科的一大研究热点，可以广泛地应用于安全部门．电视会议．身份鉴别，数字监控等领域。人脸识别技术因其广泛的应用领域而成为最具吸引力的研究方向之一。人脸是一个非常复杂的模型．因此即使一个小孩也可以轻易地识别出一张人脸．但是应用计算机进行人脸识别却是一件困难的事情。人脸识别技术有着广泛的应用前景和迫切的现实需要、是当前模式识别领域最热门的研究方向之一。一般的人脸识别系统主要包括预处理，特征提取、样本学习和识别过程四个部分．其中特征提取的好坏将直接影响到识别效果。[著者文摘]
Human face recognition is attractive in paten recognition and image processing. It can be applied to security system, human ID management, teleconference, digital surveillance and so on.The technology of face recognition is one of the most interesting research directions for its wide application. The human face is a complex paten. Though it is easy for a child to recognize a human face, it is difficult for a computer to recognize a human face automatically.Human face recognition has a wide imply respect and emergent reality need, and it is one of the most interesting research directions of mold recognition. A general system of face recognition includes： pretreatment, features extraction sample study and sample recognize. As the most important step, feature extraction will affect the rate of face recognition directly.[著者文摘]
基于神经网络的人脸特征提取及识别
於东军 杨静宇
该文使用神经网络技术对人脸的特征提取及识别做了研究。在降低图像解析度的基础上使用神经网络来进行特征压缩，可以有效地降低特征维数。使用神经网络分类器对压缩后的特征进行分类，分类结果以隶属度来表征待识别人脸属于各类别的可能性。实验结果证明了该方法的有效性。