Face recognition based on sparse manifold learning and its research

随着计算机视觉技术和机器学习技术的飞快发展，人脸识别技术应用在了生活的各个领域，例如身份识别和智能交互界面等。人脸识别是一种高维大数据集模式识别的典型应用。但数据维数过高不利于数据的分析和处理，传统的处理方法就是利用线性降维或者非线性降维的方法来解决维数灾难的问题。传统的并被广泛应用的线性降维的方法有主成分分析（PCA）,基于独立分量分析（ICA）和基于线性判别方法（LDA）等。它们在处理噪声较小的人脸识别上取得了满意的识别结果。但是人脸总会受到诸如，光照，遮盖，表情，姿势等各种噪声的影响。这样在利用线性降维的方法获取到的特征在识别上效果并不是很理想。流形学习作为近年来被提出的非线性维数约减算...Along with the rapid development of computer vision and machine learning, face recognition is used in many areas of our daily life, such as identity verification and intelligent interactive interface. Face recognition is a kind of high dimension large dataset which is typical application of pattern recognition. However, too high dimension data is not benefit for data analysis and processing, tradi...学位：工学硕士院系专业：信息科学与技术学院_计算机科学与技术学号：2302012115291