【作者】 董雅丽；

【导师】 王虹；

【作者基本信息】 武汉理工大学 ， 信号与信息处理， 2006， 硕士

【摘要】 随着社会的发展,各个方面对快速有效的自动身份验证的要求日益迫切。生物特征识别是一项利用人类特有的生理或行为特征来进行身份识别的技术,它提供了一种高可靠性、高稳定性的身份鉴别途径。而人脸识别是目前生物特征识别中最受人们关注的一个分支,是计算机视觉与模式识别领域非常活跃的一个研究方向,利用人脸特征是最自然直接的手段,相比其他生物特征,它具有直接、友好、方便的特点,易于为用户接受。在公安和安全部门有着广泛的应用,例如搜索罪犯、动态监视、银行密码系统等。人脸识别是一个涉及面广且又前景广阔的研究课题,近几年来关于人脸识别的研究取得了较大的进展。 计算机人脸识别是一个复杂而困难的问题,其原因是:人脸是由复杂的三维曲面构成的可变形体,难以用数学描述;再者,所有人的人脸结构高度相似,而人脸的图像又受年龄和成像条件的影响,这使得同一人在不同条件下的差别可能比不同的人在相同条件下的差别还要大。 因此,人脸识别技术实用化所需解决的重要问题是研究能在上述变化条件下可靠工作的人脸识别技术。 本文收集和分析了大量近年来国内外关于人脸检测与人脸识别的学术论文及研究报告,针对人脸自动识别系统的建立进行了深入的研究,提出了一种用于单人脸图像人脸识别系统的方法。实验证明本文提出的方法是合理的,有一定的理论价值与实用价值。本文的研究工作主要包括以下几个方面: 首先,在预处理环节,结合对已有算法的讨论,本文提出一种基于积分投影分析的改进算法,取得了良好的效果。 其次,在特征提取环节,重点研究了基于主成分分析(PCA)和线性判别分析(LDA)的线性特征提取方法。对二者进行比较分析的基础上,提出采用主成分分析和线性判别分析两者相结合的特征提取手段,力求达到提高识别效率的目的。在分类器设计环节,重点讨论了最近邻分类器和K—最近邻分类器。 最后,对文中讨论的上述算法的有效性,设计了一个简单的人脸识别系统,用ORL人脸像数据库进行了实验验证。结果表明本文算法设计可行,文中完成的系统也取得了较好的实验结果。更多还原

【Abstract】 With the development of the society, there are increasing demands in effective automatic identity checking. Biometrics is a kind of technology using individual physiological or behavioral characteristics to verify identity. It provides a highly reliable and high stable approach to the identity recognition. Automated face recognition is one of the most concerned branches of biometrics and it is also the one of the most active and challenging tasks for computer vision and pattern recognition at present. Others, they can be used as features for identity check, among all the characteristics of human, the characteristics of face are the most direct means which are friendly and convenient and can easily be accepted by the customers. It is widely applied to public security and security department, such as mug shots retrieval, real-time video surveillance, bank cryptography and so on. Face recognition is an extensive and challenging research problem. Recently, significant progresses have been made in the field of the face recognition technology.It is difficult to implement the face recognition mechanism using computers for several reasons. Human face is a deformable object composing of complex 3D curve surfaces, which is hard to be represented in form of mathematics. And faces of different persons have the similar structure. Furthermore, the face images are greatly dependent on ages and photography conditions. This results that the difference between two face images of two different persons taken under the same photography condition is probably less than that between two images of the same person under different conditions.Therefore, in order to implement a practical face recognition system, the recognition method needs to be independent of the above variations.The thesis collects and studies lots of papers an reports on human face detection and recognition home and abroad of these years. And it puts forward a kind of algorithm for single human face recognition. Experiments prove that the algorithm is reasonable and valuable in theory and practice. The main research work includes:Firstly, in the part of image preprocessing, it uses an advanced algorithm ofintegral projection analysis, and gains good effect.Secondly, in the part of feature extraction, it focuses on PCA and LDA. On the basis of comparison and analysis, it brings forward to associate both in order to increase the recognition ratio. In the part of classification and recognition, it discusses the Nearest Neighbor and K- Nearest Neighbor.Finally, it designs a simple face recognition system to prove the validity, and experiments on the ORL face database. The results show that the algorithm is accessible, and the system is also capable to some extent.更多还原