DC Field | Value | Language |
---|---|---|
dc.contributor.author | Park, JinSeok | ko |
dc.contributor.author | Cho, Donghyeon | ko |
dc.contributor.author | ahn, wonhyuk | ko |
dc.contributor.author | Lee, Heung-Kyu | ko |
dc.date.accessioned | 2018-11-12T04:41:36Z | - |
dc.date.available | 2018-11-12T04:41:36Z | - |
dc.date.created | 2018-10-22 | - |
dc.date.created | 2018-10-22 | - |
dc.date.issued | 2018-09-08 | - |
dc.identifier.citation | European Conf. on Computer Vision(ECCV’2018), pp.656 - 672 | - |
dc.identifier.issn | 0302-9743 | - |
dc.identifier.uri | http://hdl.handle.net/10203/246465 | - |
dc.description.abstract | Double JPEG detection is essential for detecting various image manipulations. This paper proposes a novel deep convolutional neural network for double JPEG detection using statistical histogram features from each block with a vectorized quantization table. In contrast to previous methods, the proposed approach handles mixed JPEG quality factors and is suitable for real-world situations. We collected real-world JPEG images from the image forensic service and generated a new double JPEG dataset with 1120 quantization tables to train the network. The proposed approach was verified experimentally to produce a state-of-the-art performance, successfully detecting various image manipulations. | - |
dc.language | English | - |
dc.publisher | Springer | - |
dc.title | Double JPEG Detection in Mixed JPEG Quality Factors using Deep Convolutional Neural Network | - |
dc.type | Conference | - |
dc.identifier.scopusid | 2-s2.0-85055089886 | - |
dc.type.rims | CONF | - |
dc.citation.beginningpage | 656 | - |
dc.citation.endingpage | 672 | - |
dc.citation.publicationname | European Conf. on Computer Vision(ECCV’2018) | - |
dc.identifier.conferencecountry | GE | - |
dc.identifier.conferencelocation | GASTEIG Cultural Center, Munich | - |
dc.identifier.doi | 10.1007/978-3-030-01228-1_39 | - |
dc.contributor.localauthor | Lee, Heung-Kyu | - |
dc.contributor.nonIdAuthor | Cho, Donghyeon | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.