《博士后文库》序言
Preface
Chapter 1 Introduction
1.1 Research Significance
1.2 Current Research Situation
1.2.1 Methods Based on Geometric Feature
1.2.2 Methods Based on Subspace Analysis
1.2.3 Methods Based on Machine Learning
1.2.4 Methods Based on Model
1.2.5 Methods Based on Local Feature
1.3 Classification Rules in Image Recognition
1.4 Difficulties of Face Recognition
1.5 Face Recognition System
References
Chapter 2 Image Synthesis and Classification Method Based on
Quotient Image Theory
2.1 Introduction
2.2 Background Review
2.2.1 The Quotient Image Theory
2.2.2 Illumination Subspace
2.3 The Quotient Image Method Based on 9-dimension Linear
Subspace
2.3.1 The Improved Quotient Image Method
2.3.2 Basis Image Synthesis Method
2.3.3 Illumination Direction Estimation
2.4 The Review of PCA
2.5 Face Recognition Under Different Lighting Conditions
2.6 Experiments and Results
2.6.1 The Quotient Image
2.6.2 Nine Basis Images Reconstruction
2.6.3 Face Recognition Under Varying Illumination ;
2.7 Conclusions
References
Chapter 3 A Classification Method Based on Reconstruction Error and
Normalized Distance
3.1 Introduction
3.2 Main Steps of Fusion Method Based on Reconstruction Error and
Normalized Distance
3.3 Potential Rationale of the Method
3.4 Experiments and Results
3.4.1 Experiments on the Po1yU Palmprint Database
3.4.2 Experiments on the 2D+3D Palmprint Database
3.4.3 Experiments on Corrupted Palmprint Images
3.5 Conclusions
References
Chapter 4 Integrating the Original and Approximate Face Images to
Perform Collaborative Representation Based Classification.
4.1 Introduction
4.2 Collaborative Representation Based Classification (CRC)
4.3 The Proposed Method
4.4 Experiments and Results
4.4.1 The Approximate Face Image
4.4.2 Experiments on ORL Face Database
4.4.3 Experiments on Yale Face Database
4.4.4 Experiments on FERET Face Database
4.4.5 Experiments on AR Face Database
4.5 Conclusions
References
Chapter 5 Using the Original and Symmetrical Face Training Samples to
Perform Collaborative Representation
5.1 Introduction
5.2 Collaborative Representation Based Classification(CRC)
5.3 The Proposed Method
5.4 Experiments and Results
5.4.1 The Symmetrical Face Image
5.4.2 Experiments on ORL Face Database
5.4.3 Experiments on Yale Face Database
5.4.4 Experiments on AR Face Database
5.5 Conclusions
References
Chapter 6 A Enhanced Collaborative Representation Based
Classification Method
6.1 Introduction
6.2 Collaborative Representation Based Classification (CRC)
6.3 Enhanced Collaborative Representation Based Classification
(ECRC)
6.4 Experiments and Results
6.4.1 Experiments on ORL Face Database
6.4.2 Experiments on Yale Face Database"
6.4.3 Experiments on FERET Face Database
6.5 Conclusions
References
Chapter 7 AApproximate and Competitive Representation Method with
One sample Per Person
7.1 Introduction
7.2 Main Steps of Approximate and Competitive Representation
Method
7.3 Potential Rationale of Our Method
7.4 Experiments and Results
7.4.1 Face Databases
7.4.2 Experimental Results
7.5 Conclusions
References
Chapter 8 A Kernel Twos-Phase Test Sample Sparse Representation
Method
8.1 Introduction
8.2 Two-Phase Test Sample Sparse Representation (TPTSSR)
8.3 Kernel Two-Phase Test Sample Sparse Representation
(KTPTSSR)
8.4 Experiments and Results
8.4.1 Experiments on ORL Face Database
8.4.2 Experiments on AR Face Database
8.4.3 Experiments on Yale Face Database
8.4.4 Experiments on FERET Face Database
8.5 Conclusions
References
Chapter 9 A Weighted Two-Phase Test Sample Sparse Representation
Method
9.1 Introduction
9.2 Two-Phase Test Sample Sparse Representation (TPTSSR)
9.3 Weighted Two-Phase Test Sample Sparse Representation
(WTPTSSR)
9.4 Experiments and Results

刘中华著的《复杂条件下的人脸特征提取和分类算法研究(英文版)》主要对复杂条件下的特征提取和分类算法进行了研究和总结，目标是读者可以快速了解和掌握最新的特征提取和分类算法。主要内容包括基于商图像理论的图像合成和分类方法；一种基于重构误差和均衡化距离的分类方法；整合原和近似人脸图像以及用原和对称人脸训练样本执行协同表示分类；增强协同表示分类；近似和竞争表示方法；核以及权重二阶段测试样本稀疏表示方法；基于四元数的最大边界准则方法。