返回
类型 基础研究 预答辩日期 2018-03-07
开始(开题)日期 2015-10-14 论文结束日期 2018-01-17
地点 东南大学四牌楼东南院305 论文选题来源 非立项    论文字数 7.5 (万字)
题目 单训练样本约束下人脸识别方法研究
主题词 单训练样本人脸识别,低分辨率人脸识别,人脸对称性,领域自适应,集群正则化
摘要 人脸识别是生物特征识别领域中的一个重要分支。与其他生物特征识别相比,如指纹、虹膜识别等,人脸识别具有两个重要的优势:识别的非强制性和大规模的识别能力。这意味着在人群中,即使待识别对象不配合也可以识别其身份。人脸识别的这个优势,使得人脸识别技术被广泛应用到视频监控、客流渠道控制、刑侦执法等领域。对人脸识别系统而言,人脸图像样本的采集是面临的主要难题之一。在大多数的人脸识别实际应用中,如刑侦执法,护照和身份证信息核对等,通常只能采集到每个个体的一张人脸图像,即每个人身份证件上的人脸图像,这导致了单训练样本人脸识别。许多现有的人脸识别方法在解决单训练样本人脸识别时,由于这些方法无法从单一的训练样本中提取有判别力的信息,因此它们的识别性能将会明显地降低。另外,在视频监控中,由于摄像头的安装位置与行人具有一定的距离,因而导致监控摄像头采集到的人脸图像大多是小尺寸、低分辨率的,在单训练样本约束下,如何准确地识别低分辨率人脸图像是一项更加具有挑战性的任务。本文针对上述问题,研究单训练样本约束下高、低分辨率人脸识别方法,具体研究内容如下: (1)研究基于人脸图像对称性的单训练样本人脸识别方法。利用人脸图像的对称性,扩充训练样本数量,提出多特征子空间分析方法。该方法首先把人脸图像分割成左右对称的两个半脸图像,并进一步将每张半脸图像划分成更小的人脸图像块;然后根据人脸图像块在半脸图像上的位置将它们划分成多个组,并对每一组图像块都学习一个特征子空间用以提取特征。通过学习多个特征子空间,可以有效消除人脸图像块的类间差异和类内差异之间的混淆,从而可以在每个子空间内提取更加具有判别力的特征。实验结果表明,与现有最好的方法比较,多特征子空间分析虽然原理简单,但是无论识别约束环境下还是无约束环境下拍摄的人脸图像,都能够实现更高或者相差无几的识别准确率。 (2)研究基于领域自适应的单训练样本人脸识别方法。为充分利用已建立的多训练样本人脸数据库解决单训练样本人脸识别问题,提出增强判别特征学习方法。该方法假设多训练样本数据集和单训练样本数据集/测试集分别来自两个具有不同数据分布的领域(分别称为源领域和目标领域),并且从领域自适应的角度构建优化模型,其目标是将从源领域学习的判别模型迁移到目标领域上执行分类任务。考虑到目标领域训练集包含的样本数量太少,无法准确表示目标领域的数据分布,首先提出局部保持领域自适应算法以实现良好的领域适应效果;其次,为获得最有利于判别模型学习的领域自适应特征,将领域自适应和判别模型学习统一到同一个学习框架,并在判别模型学习中考虑特征选择来提高其对目标领域的噪声和异常样本的鲁棒性。对增强判别特征学习方法在以下两种情形下进行评估:不同种族人脸图像之间的领域自适应和不同图像采集条件下人脸图像之间的领域自适应,实验结果表明该方法可以有效解决单训练样本人脸识别。 (3)研究基于集群正则化的单训练样本低分辨率人脸识别方法。在低分辨率人脸识别中,针对单训练样本导致的奇异矩阵和模型过拟合问题,提出基于集群正则化的同步判别分析方法。由于聚类算法可以将样本按照相似度划分为多个集群,基于集群正则化的同步判别分析利用对高分辨率人脸图像的聚类结果,分别使用集群间散度矩阵和集群内散度矩阵来正则化类间散度矩阵和类内散度矩阵。通过使用基于集群的散度矩阵,不仅可以解决模型求解中遇到的矩阵奇异性问题,而且可以解决模型学习的过拟合问题。在单训练样本约束下,与现有的低分辨率人脸识别方法比较,基于集群正则项的同步判别分析方法在识别约束环境下和无约束环境下拍摄的低分辨率人脸图像时都可以获得更好的识别结果。
英文题目 Research on Face Recognition Methods For Single Sample per Person in the Training Set
英文主题词 Single sample per person face recognition, low-resolution face recognition, facial symmetry, domain adaptation,cluster-based regularization
英文摘要 Face recognition is an important branch in the field of biometric recognition. Compared with other biometric recognition, such as fingerprint and iris recognition, face recognition has two important advantages: non-compulsory recognition and large scale recognition. This means that a person can be identified by face recognition in the crowd, even if he or she is not cooperative. These advantages of face recognition make face recognition technology widely used in video surveillance, access control, law enforcement, etc. Collecting of face samples is one of the main difficulties for face recognition. In most of the real-world applications such as law enhancement, e-passport, and ID card identification, it is customary to collect a single sample per person (SSPP), i.e., the profile face image, which leads to SSPP face recognition problem. Unfortunately, in such SSPP scenario, many presented face recognition methods suffer serious performance drop or fail to work due to their inability to learn the discriminative information of a person from a single sample. Moreover, in video surveillance, as the installed cameras are far away from the pedestrians, face images captured by these surveillance cameras tend to be with small size and low-resolution. In SSPP scenario, how to accurately recognize the low-resolution faces is a more challenging task. Aiming at sovling the above problems, both the high-resolution and the low-resolution face recognition for single sample per person are investigated in this dessertation. In detail, the contents are as follows: (1) Based on facial symmetry, a multiple feature subspace analysis (MFSA) approach is proposed for SSPP face recognition. In MFSA, the facial symmetry is used to augment the number of samples in the training set. First, MFSA divides each enrolled face into two halves about the bilateral symmetry axis, and further partition every half into several local face patches. Second, it clusters all the patches into multiple groups according to their locations at the half face and learns a feature subspace for each group of patches. By learning multiple feature subspace, the confusion between inter-class and intra-class variations of face patches is removed and more discriminative features can be extracted from each subspace. Compared with the state-of-the-art methods, MFSA is effortless and efficient in implementing, but achieves either better or competitive performance when recognizing face images taken in both constrained and unconstrained environment. (2) Based on domain adaptation, an enhanced discriminative feature learning (EDFL) method is developed to deal with SSPP face recognition. EDFL is designed to make full use of existing face databases (named auxiliary dataset) which includes multiple face sample per person. EDFL treats the auxiliary dataset and gallery/probe dataset come from two domains (named source domain and target domain, respectively) with different data distributions, it is modeled from the perspective of domain adaptation. The aim of EDFL is to transfer the discriminative knowledge learned from the source domain to the target domain for classification. Firstly, considering that the galley set in target domain contains too limited sample to accurately represent the data distribution of the target domain, locality preserving domain adaption (LPDA) is presented to realize good overall domain adaptation. Secondly, to guarantee the learned domain adaptation components are optimal for discriminative learning, EDFL unifies the domain adaptation and discriminant model learning into the same framework. Meanwhile, to make the learned discriminant model robust to noises and outliers in the target domain, feature selection is taken into account in the model learning. EDFL is extensively evaluated under two scenarios, i.e., enhanced discriminative feature learning across ethnicity and across imaging condition. The comparison results with some state-of-the-art methods demonstrate the effectiveness of EDFL on SSPP face recognition. (3) A cluster-based regularized simultaneous discriminant analysis (C-RSDA) method is developed to address low-resolution face recognition with SSPP. For low-resolution face recognition in SSPP scenario, it often suffers from overfitting and singular matrix problems. Because the clustering algorithms are able to gather similar samples while separate dissimilar ones, C-RSDA regularizes the between-class and within-class scatter matrices respectively with inter-cluster and intra-cluster scatter matrices, where the cluster-based scatter matrices are computed according to the clustering results of high-resolution face images. With the cluster-based scatter matrices, not only the singularity problem is resolved, but overfitting problem is overcomed as more variations are exploited from the limited training samples. C-RSDA is extensively evaluated on recognizing low-resolution face images captured in both controlled and uncontrolled environments. Compared with exiting methods designed for low-resoution face recognition, the encouraging experimental results of C-RSDA demonstrate its superiority over other methods.
学术讨论
主办单位时间地点报告人报告主题
导师课题组 2014.07.24 经管楼B601 楚永杰 基于白化特征的二维线性判别分析
导师课题组 2014.12.10 经管楼B601 楚永杰 Multiple Feature Subspaces Analysis for Single Sample per Person Face Recognition
宁波大学 2015.9.12 宁波大学商学院 楚永杰 Enhanced Discriminative Subspace Learning via Domain Adaptation and Feature Selection
东南大学系统工程研究所 2013.10.23 经管楼B201 张娟 Ingredient branding strategies in an assembly supply chain: models and analysis
东南大学系统工程研究所 2013.10.29 经管楼B201 Yufei Yuan Logistics for Large-scale Disaster Response Achievements and Challenges
东南大学管理科学与工程系 2014.06.10 经管楼B201 Kim-Chuan Toh Algorithm for large-scale matrix optimization
东南大学系统工程研究所 2014.06.24 经管楼B201 Pang Zhan Coordinating Inventory Control and Pricing strategies for Perishable Products
导师课题组 2013.12.24 经管楼B601 楚永杰 Face Recognition Using Fisher Vectors
     
学术会议
会议名称时间地点本人报告本人报告题目
第十届物流系统工程学术研讨会 2014.10.26-2017.10.28 大连理工大学盘锦校区 Two-dimensional Linear Discriminant Analysis based on Whitening Features
第十二届创新计算、信息和控制国际会议 2017.08.28-2017.08.30 日本久留米市 A novel log-based weighted 2DLDA algorithm for face recognition
     
代表作
论文名称
Low-resolution face recognition with single sample per person
 
答辩委员会组成信息
姓名职称导师类别工作单位是否主席备注
张智光 正高 教授 博导 南京林业大学
徐小林 正高 教授 博导 南京大学
王海燕 正高 教授 博导 东南大学
何勇 正高 教授 博导 东南大学
李四杰 副高 副教授 博导 东南大学
      
答辩秘书信息
姓名职称工作单位备注
薛巍立 副高 副教授 东南大学