CASIA OpenIR  > 毕业生  > 博士学位论文
面向生物识别的特征选择与学习方法研究
其他题名Feature selection and learning for biometrics
王立彬
学位类型工学博士
导师谭铁牛
2015-05-17
学位授予单位中国科学院大学
学位授予地点中国科学院自动化研究所
学位专业模式识别与智能系统
关键词生物识别 特征选择 稀疏学习 虹膜识别 定序测量 特征映射 Biometrics Feature Selection Sparse Learning Iris Recognition Ordinal Measures Feature Mapping
摘要随着信息技术的智能化发展,基于生物特征识别的身份认证以其唯一性、稳定性的特性,和灵活便捷的使用方式得到广泛应用,其中基于图像的生物特征识别技术又具有显著的优势,包括虹膜识别、掌纹识别、人脸识别等。在一个完整的生物识别系统的核心算法中,生物特征图像的特征分析至关重要。特征分析旨在设计或学习得到一组紧致而且有区分力的滤波器,能够提取图像鲁棒的本质属性信息,这其中主要涉及到特征描述子的构造,特征选择,特征映射等学习过程。 本文将围绕多模态视觉生物识别系统,利用机器学习算法,在图像的局部特征描述,特征选择,特征映射学习等流程开展以下工作: 1.针对局部特征描述子的优化,总结了定序测量特征参数变化对纹理图像识别的影响规律,在空间差分滤波器的基础上,提出面向图像内容的纹理滤波器,同时利用空间定序测量特征和纹理定序测量特征的优势,提升了特征表达的鲁棒性和可区分力。 2.提出了基于鲁棒正则化线性规划模型的特征选择方法,是一种全局优化的稀疏学习模型,其特性为采用最大间隔的损失函数,引入特征分类能力的先验信息,并附加非负的稀疏约束。大规模的虹膜识别和掌纹识别实验验证了该模型的分类性能和泛化性。 3.提出了基于后向反馈稀疏学习模型的固定数目特征选择方法,归纳一类启发式的正则化稀疏学习模型的一般算法框架,并同时提出了鲁棒浮动稀疏学习和非负浮动稀疏学习两种扩展模型,实现了高效鲁棒的特征选择。 4. 提出了基于类Boosting策略的稀疏化特征选择方法,通过改变样本权重分布选择出不同类型的特征。并结合样本选择方法构成 两阶段的学习方法,能够高效的得到稀疏并且具有互补性的紧致特征集合。 5. 提出了基于分段线性的特征映射和特征选择联合学习方法,通过显示映射将低维特征投影到高维特征空间,并结合单特征和成对特征 的映射策略,充分挖掘所选择特征的分类能力。 综上所述,我们以生物特征图像识别为主线,深入分析其核心特征分析中各个环节的问题,并提出了相应的有效解决方案,有助于提升实际 多模态生物识别身份认证系统的性能。
其他摘要Biometrics is widely used in a flexible and convenient way, because of its uniqueness and stability, This is especially true for image-based biometrics, such as iris recognition, palmprint recognition, and face recognition. Feature representation of biometric images plays an important role in a real biometric system. Feature representation aims to design or learn a set of compact and discriminative filters which can extract robust and essential characters from the images. It involves the design of feature descriptors, feature selection and feature mapping. This thesis focuses on multimodal biometric systems to develop related machine learning algorithms on local feature extraction, feature selection and feature mapping learning. Main contributions of the thesis are as follows: 1.For feature descriptor, the influence of parameters of ordinal measures is studied and summarized for texture image recognition. Image content based texture filter is proposed except for spatial differential filter.Combining the spatial ordinal measures with texture ordinal measures can improve the robustness and discrimination of features. 2. A robust regularized linear programming feature selection model is proposed as a global optimal sparse learning model. The large margin criterion is adopted to design the loss function, and the useful prior information of each feature is introduced to update the weight. Finally a non-negative sparse constraint is enforced to ensure the sparsity of features.Extensive experiments on iris and palmprint recognition are conducted to validate the effectiveness and generalization of the proposed algorithm. 4. A back-forward sparse learning algorithm is proposed to select the top k features, and a unified framework of heuristic regularized sparse learning algorithm is summarized. Two extensions of robust FloatSparse and nonnegative FloatSparse learning model are formulated to implement efficient and robust top k feature selection. 5. A Boosting-like sparsity regularized model is proposed with sample selection. The weights of samples are updated at each iteration, thus different styles of features are selected at different iterations. Finally the two-stage learning process can obtain both sparse and complementary features. 6. A joint piecewise linear feature mapping and selection algorithm is proposed to map the selected low-dimensional features to high-dimensional features. Through combining single feature mapping and pairwise feature mapping, ...
其他标识符201218014628061
语种中文
文献类型学位论文
条目标识符http://ir.ia.ac.cn/handle/173211/6668
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
王立彬. 面向生物识别的特征选择与学习方法研究[D]. 中国科学院自动化研究所. 中国科学院大学,2015.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
CASIA_20121801462806(11629KB) 暂不开放CC BY-NC-SA请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[王立彬]的文章
百度学术
百度学术中相似的文章
[王立彬]的文章
必应学术
必应学术中相似的文章
[王立彬]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。