Graph Matching Using a Direct Classification of Node Attendance.pdfVIP

下载本文档

4
0
约6.19万字
约 29页
2017-04-13 发布于江苏
举报
版权申诉

Graph Matching Using a Direct Classification of Node Attendance.pdf

1、本文档共29页，可阅读全部内容。
2、原创力文档（book118）网站文档一经付费（服务费），不意味着购买了该文档的版权，仅供个人/单位学习、研究之用，不得用于商业用途，未经授权，严禁复制、发行、汇编、翻译或者网络传播等，侵权必究。
3、本站所有内容均由合作方或网友上传，本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺！文档内容仅供研究参考，付费前请自行鉴别。如您付费，意味着您自己接受本站规则且自行承担风险，本站不退款、不进行额外附加服务；查看《如何避免下载的几个坑》。如果您已付费下载过本站文档，您可以点击这里二次下载。
4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等，请点击“版权申诉”（推荐），也可以打举报电话：400-050-0827(电话支持时间：9:00-18:30)。
5、该文档为VIP文档，如果想要下载，成为VIP会员后，下载免费。
6、成为VIP后，下载本文档将扣除1次下载权益。下载后，不支持退款、换文档。如有疑问请联系我们。
7、成为VIP后，您将拥有八大权益，权益包括：VIP文档下载权益、阅读免打扰、文档格式转换、高级专利检索、专属身份标志、高级客服、多端互通、版权登记。
8、VIP文档为合作方或网友上传，每下载1次，网站将根据用户上传文档的质量评分、类型等，对文档贡献者给予高额补贴、流量扶持。如果你也想贡献VIP文档。上传文档

Graph Matching Using a Direct Classification of Node Attendance

Graph Matching Usinga Direct Classication of Node Attendance Fred DePiero and Mohan TrivediComputer Vision and Robotics Research LaboratoryElectrical and Computer Engineering DepartmentThe University of Tennessee, Knoxville, TN 37996-2100depiero@trivedi@Steve SerbinMathematics DepartmentUniversity of Tennessee, Knoxville, TN 37996-2100serbin@AbstractAn algorithm has been developed that nds isomorphisms between both graphs and subgraphs. Thedevelopment is introduced in the object recognition problem domain. The method isolates matchingsubgraphs, nds a node-to-node mapping and reorders nodes thus permitting a direct comparison tobe made between the resultant graphs. The algorithm is of polynomial order. It yields approximateresults, maintaining a performance level for subgraph isomorphisms at or above 95% under a widevariety of conditions and with varying levels of noise. The performance on the full size comparisonsassociated with graph isomorphisms has been found to be 100/100, also under a variety of conditions.Performance metrics, methods of testing and results are presented.KEYWORDS: Direct Classication, Graph Isomorphism, Subgraph Isomorphism,Graph Matching, Object Recognition.1 IntroductionObject recognition is fundamentally a problem of subgraph isomorphism in that a model describesobjects in their entirety; in contrast to an observed object where all features are not typically seen ina single view [1]. Current techniques using range data for shape-based recognition of cluttered scenesare typically quite time consuming and can have undesirable tradeos between speed and accuracy.0This research is sponsored in part by a grant awarded by the Oce of Technology Development, U.S. Department ofEnergy as part of the Environmental Restoration and Waste Management Program. Techniques are also sometimes tailored for the bin picking problem which focuses on a cluttered sceneinvolving just a few dierent kinds of objects [2]. We pursue active vision research [3] top