A new watermarking method with obfuscated quasi-chirp transform

Kazuo Ohzeki; Yuanyu Wei; Yutaka Hirakawa; Kiyotsugu Sato

doi:10.1007/978-3-642-32205-1_7

A new watermarking method with obfuscated quasi-chirp transform

Kazuo Ohzeki, Yuanyu Wei, Yutaka Hirakawa, Kiyotsugu Sato

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

1 Citation (Scopus)

Abstract

Watermark detection software is obfuscated using a table to hide embedding and detection algorithms. As the table size is limited, the block size is also limited for watermarking. To address this situation, a new quasi-chirp transform is developed to improve embedding efficiency. The quasi-chirp transform is different from the conventional DCT or Fourier transform. It contains multiple frequency components in a single basis of the transform. It disperses image data rather than compressing it, as the DCT does. The dispersed data increases the range for embedding watermarks. The chirp transform is able to embed even on a flat area of an image. Using this chirp transform, embedding and detection experiments for image data with small block sizes were carried out. A high SNR and robust watermark with an evaluated obfuscation were obtained.

Original language	English
Title of host publication	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages	57-71
Number of pages	15
Volume	7128 LNCS
DOIs	https://doi.org/10.1007/978-3-642-32205-1_7
Publication status	Published - 2012
Event	10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011 - Atlantic City, NJ Duration: 2011 Oct 23 → 2011 Oct 26

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	7128 LNCS
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011
City	Atlantic City, NJ
Period	11/10/23 → 11/10/26

Fingerprint

Keywords

detection
embedding
obfuscation
orthogonal transform
quantization

ASJC Scopus subject areas

Computer Science(all)
Theoretical Computer Science

Cite this

Ohzeki, K., Wei, Y., Hirakawa, Y., & Sato, K. (2012). A new watermarking method with obfuscated quasi-chirp transform. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7128 LNCS, pp. 57-71). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7128 LNCS). https://doi.org/10.1007/978-3-642-32205-1_7

A new watermarking method with obfuscated quasi-chirp transform. / Ohzeki, Kazuo; Wei, Yuanyu; Hirakawa, Yutaka; Sato, Kiyotsugu.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7128 LNCS 2012. p. 57-71 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7128 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Ohzeki, K, Wei, Y, Hirakawa, Y & Sato, K 2012, A new watermarking method with obfuscated quasi-chirp transform. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 7128 LNCS, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 7128 LNCS, pp. 57-71, 10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011, Atlantic City, NJ, 11/10/23. https://doi.org/10.1007/978-3-642-32205-1_7

Ohzeki K, Wei Y, Hirakawa Y, Sato K. A new watermarking method with obfuscated quasi-chirp transform. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7128 LNCS. 2012. p. 57-71. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-32205-1_7

Ohzeki, Kazuo ; Wei, Yuanyu ; Hirakawa, Yutaka ; Sato, Kiyotsugu. / A new watermarking method with obfuscated quasi-chirp transform. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 7128 LNCS 2012. pp. 57-71 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{5837242dd07c42c692a8aa495ea1d92d,

title = "A new watermarking method with obfuscated quasi-chirp transform",

abstract = "Watermark detection software is obfuscated using a table to hide embedding and detection algorithms. As the table size is limited, the block size is also limited for watermarking. To address this situation, a new quasi-chirp transform is developed to improve embedding efficiency. The quasi-chirp transform is different from the conventional DCT or Fourier transform. It contains multiple frequency components in a single basis of the transform. It disperses image data rather than compressing it, as the DCT does. The dispersed data increases the range for embedding watermarks. The chirp transform is able to embed even on a flat area of an image. Using this chirp transform, embedding and detection experiments for image data with small block sizes were carried out. A high SNR and robust watermark with an evaluated obfuscation were obtained.",

keywords = "detection, embedding, obfuscation, orthogonal transform, quantization",

author = "Kazuo Ohzeki and Yuanyu Wei and Yutaka Hirakawa and Kiyotsugu Sato",

year = "2012",

doi = "10.1007/978-3-642-32205-1_7",

language = "English",

isbn = "9783642322044",

volume = "7128 LNCS",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "57--71",

booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

note = "10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011 ; Conference date: 23-10-2011 Through 26-10-2011",

}

TY - GEN

T1 - A new watermarking method with obfuscated quasi-chirp transform

AU - Ohzeki, Kazuo

AU - Wei, Yuanyu

AU - Hirakawa, Yutaka

AU - Sato, Kiyotsugu

PY - 2012

Y1 - 2012

N2 - Watermark detection software is obfuscated using a table to hide embedding and detection algorithms. As the table size is limited, the block size is also limited for watermarking. To address this situation, a new quasi-chirp transform is developed to improve embedding efficiency. The quasi-chirp transform is different from the conventional DCT or Fourier transform. It contains multiple frequency components in a single basis of the transform. It disperses image data rather than compressing it, as the DCT does. The dispersed data increases the range for embedding watermarks. The chirp transform is able to embed even on a flat area of an image. Using this chirp transform, embedding and detection experiments for image data with small block sizes were carried out. A high SNR and robust watermark with an evaluated obfuscation were obtained.

AB - Watermark detection software is obfuscated using a table to hide embedding and detection algorithms. As the table size is limited, the block size is also limited for watermarking. To address this situation, a new quasi-chirp transform is developed to improve embedding efficiency. The quasi-chirp transform is different from the conventional DCT or Fourier transform. It contains multiple frequency components in a single basis of the transform. It disperses image data rather than compressing it, as the DCT does. The dispersed data increases the range for embedding watermarks. The chirp transform is able to embed even on a flat area of an image. Using this chirp transform, embedding and detection experiments for image data with small block sizes were carried out. A high SNR and robust watermark with an evaluated obfuscation were obtained.

KW - detection

KW - embedding

KW - obfuscation

KW - orthogonal transform

KW - quantization

UR - http://www.scopus.com/inward/record.url?scp=84865682794&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84865682794&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-32205-1_7

DO - 10.1007/978-3-642-32205-1_7

M3 - Conference contribution

AN - SCOPUS:84865682794

SN - 9783642322044

VL - 7128 LNCS

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 57

EP - 71

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

T2 - 10th International Workshop on Digital-Forensics and Watermarking, IWDW 2011

Y2 - 23 October 2011 through 26 October 2011

ER -

Access to Document

10.1007/978-3-642-32205-1_7