近红外人脸图像的分类研究.doc
约37页DOC格式手机打开展开
近红外人脸图像的分类研究,1.24万字我自己的毕业论文，原创的，已经通过校内系统检测，仅在本站独家出售，重复率低，大家放心下载使用摘要 近年来人们越来越多的关注生物特征识别，而在这些生物特征识别方法中，人脸识别具有方便，经济而准确的特点，可以广泛应用于高安全性部门的警戒、入口控制、人机交互、计算机保密、公共安全等方面。在...
内容介绍
此文档由会员 淘宝大梦 发布
近红外人脸图像的分类研究
1.24万字
我自己的毕业论文，原创的，已经通过校内系统检测，仅在本站独家出售，重复率低，大家放心下载使用
摘要 近年来人们越来越多的关注生物特征识别，而在这些生物特征识别方法中，人脸识别具有方便，经济而准确的特点，可以广泛应用于高安全性部门的警戒、入口控制、人机交互、计算机保密、公共安全等方面。在人脸识别中，人脸外观会受到光照、姿态、表情变化的影响，所以人脸识别系统要适应各种环境，此外化妆、照片欺诈也是人脸识别亟待解决的问题。本文主要研究基于二维线性判别分析的人脸识别。
本文首先介绍人脸识别的研究意义、识别方法、研究中的问题和难点以及人脸识别的现状和发展前景。研究以近红外图像为基础的人脸识别方法，并且也了解近红外图像人脸识别的特点和价值，它的缺点和可提高识别性能的研究方向也是要了解的。根据以上内容，本文着重研究二维线性判别分析（2DLDA）方法。2DLDA算法识别率高、处理速度快，可以对人脸图像数据降维，将人脸图像测试样本变换为二维矩阵并进行列/行方向的二维线性判别分析。这是2DLDA算法的优势。
在Matlab上对基于二维线性判别分析的人脸识别算法进行编程实现。
关键词 近红外 人脸识别 二维线性判别
Classification of near infrared face image
Abstract In recent years, more and more people were concerned about biometric identification, and in these biometric identification methods, face recognition with the characteristic of convenient, economical and accurate, can be widely used in high security alert authorities, access control, human interactive, computer secrecy, public safety and other aspects. In face recognition,light, gesture, facial expression affected face appearance, so the face recognition system have to adapt to the environment. In addition to makeup, photo recognition fraud were also serious problems. This paper studied the research of face recognition based on two-dimensional linear discriminant analysis.
This paper introduced the significance of face recognition, identification methods, research problems and difficulties as well as research status of face recognition and development prospects. The paper discussed the characteristics and values of face recognition based on the near-infrared image , comprehended the face recognition methods based on near-infrared image, meanwhile discussed the shortcomings of face recognition research based on the near-infrared image and improve recognition performance. On this basis, the paper focused on the two-dimensional linear discriminant analysis (TDLDA) method. 2DLDA algorithm had higher recognition rate, fast processing speed, can reduce the dimensionality of face image data, the face image test samples were converted into a two-dimensional matrix and thus made two-dimensional linear discriminant analysis of row/column direction. This is the advantages of 2DLDA algorithm.
A program about the algorithm of face recognition based on 2DLDA is written on MATLAB.
Key words Near-infrared facial recognition 2DLDA
目录
第一章 绪论…………………………………………………………………………........1
1.1 研究人脸识别的意义……………………………………………..............................1
1.2 人脸识别的研究内容……………………………………………..............................1
1.3 人脸识别研究现状及难点………………………………………..............................1
1.4 人脸识别系统……………………………………………………..............................2
1.5 人脸识别的发展趋势及应用领域………………………………..............................2
1.6 本文主要工作及结构安排……………………………………………………...….3
第二章 近红外人脸识别及二维线性判别分析………..………............................ 4
2.1 近红外人脸识别……………………………………………………………...….....4
2.1.1 近红外人脸识别的意义……………………………………………………...……4
2.1.2 近红外图像…………………………………………………………………...……4
2.2 人脸图片库……………………………………………………………………...….5
2.3 线性判别分析…………………………................ ………………...........................6
2.3.1 LDA与2DLDA………………………………………………………………........6
2.3.2 线性判别分析的缺点与改进………………………………………………….......6
第三章 基于二维线性判别分析的人脸识别………………………………….…...8
3.1 线性判别分析的原理…………………………………………………………..........8
3.2 线性判别法……………………………………………………………………...….9
3.3 2DPCA原理……………………………………………………………...………….10
3.4 2DLDA原理……………………………………………………………...………….10
3.5 2DLDA在人脸识别中的应用…………………………………….…………...……12
3.5.1 特征提取……………………………………………………………………….....12
3.5.2 分类…………………………………………………………………………..…...12
3.6 算法评价标准…………………………………………………………………….....13
第四章 MATLAB编程实现及运行结果分析……………………………………... 14
4.1 MATLAB相关函数简介………..
1.24万字
我自己的毕业论文，原创的，已经通过校内系统检测，仅在本站独家出售，重复率低，大家放心下载使用
摘要 近年来人们越来越多的关注生物特征识别，而在这些生物特征识别方法中，人脸识别具有方便，经济而准确的特点，可以广泛应用于高安全性部门的警戒、入口控制、人机交互、计算机保密、公共安全等方面。在人脸识别中，人脸外观会受到光照、姿态、表情变化的影响，所以人脸识别系统要适应各种环境，此外化妆、照片欺诈也是人脸识别亟待解决的问题。本文主要研究基于二维线性判别分析的人脸识别。
本文首先介绍人脸识别的研究意义、识别方法、研究中的问题和难点以及人脸识别的现状和发展前景。研究以近红外图像为基础的人脸识别方法，并且也了解近红外图像人脸识别的特点和价值，它的缺点和可提高识别性能的研究方向也是要了解的。根据以上内容，本文着重研究二维线性判别分析（2DLDA）方法。2DLDA算法识别率高、处理速度快，可以对人脸图像数据降维，将人脸图像测试样本变换为二维矩阵并进行列/行方向的二维线性判别分析。这是2DLDA算法的优势。
在Matlab上对基于二维线性判别分析的人脸识别算法进行编程实现。
关键词 近红外 人脸识别 二维线性判别
Classification of near infrared face image
Abstract In recent years, more and more people were concerned about biometric identification, and in these biometric identification methods, face recognition with the characteristic of convenient, economical and accurate, can be widely used in high security alert authorities, access control, human interactive, computer secrecy, public safety and other aspects. In face recognition,light, gesture, facial expression affected face appearance, so the face recognition system have to adapt to the environment. In addition to makeup, photo recognition fraud were also serious problems. This paper studied the research of face recognition based on two-dimensional linear discriminant analysis.
This paper introduced the significance of face recognition, identification methods, research problems and difficulties as well as research status of face recognition and development prospects. The paper discussed the characteristics and values of face recognition based on the near-infrared image , comprehended the face recognition methods based on near-infrared image, meanwhile discussed the shortcomings of face recognition research based on the near-infrared image and improve recognition performance. On this basis, the paper focused on the two-dimensional linear discriminant analysis (TDLDA) method. 2DLDA algorithm had higher recognition rate, fast processing speed, can reduce the dimensionality of face image data, the face image test samples were converted into a two-dimensional matrix and thus made two-dimensional linear discriminant analysis of row/column direction. This is the advantages of 2DLDA algorithm.
A program about the algorithm of face recognition based on 2DLDA is written on MATLAB.
Key words Near-infrared facial recognition 2DLDA
目录
第一章 绪论…………………………………………………………………………........1
1.1 研究人脸识别的意义……………………………………………..............................1
1.2 人脸识别的研究内容……………………………………………..............................1
1.3 人脸识别研究现状及难点………………………………………..............................1
1.4 人脸识别系统……………………………………………………..............................2
1.5 人脸识别的发展趋势及应用领域………………………………..............................2
1.6 本文主要工作及结构安排……………………………………………………...….3
第二章 近红外人脸识别及二维线性判别分析………..………............................ 4
2.1 近红外人脸识别……………………………………………………………...….....4
2.1.1 近红外人脸识别的意义……………………………………………………...……4
2.1.2 近红外图像…………………………………………………………………...……4
2.2 人脸图片库……………………………………………………………………...….5
2.3 线性判别分析…………………………................ ………………...........................6
2.3.1 LDA与2DLDA………………………………………………………………........6
2.3.2 线性判别分析的缺点与改进………………………………………………….......6
第三章 基于二维线性判别分析的人脸识别………………………………….…...8
3.1 线性判别分析的原理…………………………………………………………..........8
3.2 线性判别法……………………………………………………………………...….9
3.3 2DPCA原理……………………………………………………………...………….10
3.4 2DLDA原理……………………………………………………………...………….10
3.5 2DLDA在人脸识别中的应用…………………………………….…………...……12
3.5.1 特征提取……………………………………………………………………….....12
3.5.2 分类…………………………………………………………………………..…...12
3.6 算法评价标准…………………………………………………………………….....13
第四章 MATLAB编程实现及运行结果分析……………………………………... 14
4.1 MATLAB相关函数简介………..
