DC Field | Value | Language |
---|---|---|
dc.contributor.author | Lee, SeongHun | - |
dc.contributor.author | Cho, Min Su | - |
dc.contributor.author | Jung, Kyomin | - |
dc.contributor.author | Kim, Jin Hyung | - |
dc.date.accessioned | 2011-03-24T04:47:57Z | - |
dc.date.available | 2011-03-24T04:47:57Z | - |
dc.date.issued | 2011-03-24 | - |
dc.identifier.uri | http://hdl.handle.net/10203/22972 | - |
dc.description.abstract | In this paper, we propose a framework for isolating text regions from natural scene images. The main algorithm has two functions: it generates text region candidates, and it verifies of the label of the candidates (text or non-text). The text region candidates are generated through a modified Kmeans clustering algorithm, which references texture features, edge information and color information. The candidate labels are then verified in a global sense by the Markov Random Field model where collinearity weight is added as long as most texts are aligned. The proposed method achieves reasonable accuracy for text extraction from moderately dicult examples from the ICDAR 2003 database. | en |
dc.description.sponsorship | This work was supported by the KOSEF grant funded by the Korea government 2009-0078943, BK21 and NAP of Korea Research Council of Fundamental Science & Technology | en |
dc.language.iso | en_US | en |
dc.title | Scene Text Extraction with Edge Constraint and Text Collinearity | en |
dc.type | Article | en |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.