Show simple item record

dc.contributor.advisor丁兴号
dc.contributor.author糜增元
dc.date.accessioned2016-01-13T09:08:56Z
dc.date.available2016-01-13T09:08:56Z
dc.date.issued2013-12-16 15:36:56.0
dc.identifier.urihttps://dspace.xmu.edu.cn/handle/2288/79474
dc.description.abstract随着信息技术的不断发展,人们获取的信息量日益膨胀,如何对信号进行简洁有效的表示已经成为研究热点之一。稀疏表示作为一种新兴的信号建模方式,能够以较少的非零系数有效提取信号的本质特性,减少所需要处理的数据量,已在信号去噪、视频压缩、模式识别等领域得到广泛应用。过完备字典的构建以及稀疏分解算法是稀疏表示理论的关键。传统的稀疏表示模型通常是建立在观测数据所含噪声服从高斯分布的假设前提,但是当观测数据混有野点噪声(例如图像中的椒盐噪声)时,这样的假设往往会导致不精确的重建结果。 本文在深入研究稀疏表示理论基础上,针对现有稀疏表示算法缺陷,结合稀疏贝叶斯学习理论,提出两种基于非参数贝叶斯框架的鲁棒稀疏表...
dc.description.abstractWith the improvement of information technology, how to represent a signal briefly and effectively has become a research focus nowadays. Sparse representation ,which is a new signal representation theory,has been widely used in fields such as signal de-noising, video compression and pattern recognition. Only a few number of nonzero coefficients are needed to reveal essential features of the signal ...
dc.language.isozh_CN
dc.relation.urihttps://catalog.xmu.edu.cn/opac/openlink.php?strText=38902&doctype=ALL&strSearchType=callno
dc.source.urihttps://etd.xmu.edu.cn/detail.asp?serial=39050
dc.subject稀疏表示
dc.subject自适应字典学习
dc.subject非参数贝叶斯
dc.subject野点
dc.subjectSparse Representation
dc.subjectAdaptive Dictionary Learning
dc.subjectNon-parametric Bayesian Framework
dc.subjectOutlier
dc.title基于贝叶斯框架的鲁棒自适应字典稀疏表示理论及应用研究
dc.title.alternativeSparse Representation based on Adaptive Dictionary Learning under Bayesian Framework and the Research of Application
dc.typethesis
dc.date.replied2013-06-02
dc.description.note学位:工学硕士
dc.description.note院系专业:信息科学与技术学院_信号与信息处理
dc.description.note学号:23320101153150


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record