File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Document image watermarking based on weight-invariant partition using support vector machine

TitleDocument image watermarking based on weight-invariant partition using support vector machine
Authors
Issue Date2004
Citation
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2004, v. 3163, p. 546-554 How to Cite?
AbstractSecurity concern about document images is important since document images are distributed in large amount both electronically and physically. They are easily copied and the copyright information is difficult to identify. In this paper, we present a new algorithm for document image watermarking, which is based on weight-invariant partition of the document in the spatial domain, where the weight of a partition is the average number of pixels in lines within the partition. We discuss the issues of robustness, security, capacity and fault-tolerance related to the watermarking method. In order to simultaneously achieve high capacity and security, the partition method is further improved using the support vector machine technique. The experiments indicate the soundness of our method. © Springer-Verlag 2004.
Persistent Identifierhttp://hdl.handle.net/10722/336059
ISSN
2023 SCImago Journal Rankings: 0.606

 

DC FieldValueLanguage
dc.contributor.authorHu, Shiyan-
dc.date.accessioned2024-01-15T08:22:26Z-
dc.date.available2024-01-15T08:22:26Z-
dc.date.issued2004-
dc.identifier.citationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2004, v. 3163, p. 546-554-
dc.identifier.issn0302-9743-
dc.identifier.urihttp://hdl.handle.net/10722/336059-
dc.description.abstractSecurity concern about document images is important since document images are distributed in large amount both electronically and physically. They are easily copied and the copyright information is difficult to identify. In this paper, we present a new algorithm for document image watermarking, which is based on weight-invariant partition of the document in the spatial domain, where the weight of a partition is the average number of pixels in lines within the partition. We discuss the issues of robustness, security, capacity and fault-tolerance related to the watermarking method. In order to simultaneously achieve high capacity and security, the partition method is further improved using the support vector machine technique. The experiments indicate the soundness of our method. © Springer-Verlag 2004.-
dc.languageeng-
dc.relation.ispartofLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)-
dc.titleDocument image watermarking based on weight-invariant partition using support vector machine-
dc.typeArticle-
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1007/978-3-540-28640-0_51-
dc.identifier.scopuseid_2-s2.0-35048830484-
dc.identifier.volume3163-
dc.identifier.spage546-
dc.identifier.epage554-
dc.identifier.eissn1611-3349-

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats