Enhanced Perceptual Image Authentication with Tamper Localization and Self-Restoration

Liu, F, Wang, H, Cheng, L-M, Ho, ATS and Li, S (2014) Enhanced Perceptual Image Authentication with Tamper Localization and Self-Restoration In: 2014 IEEE International Conference on Multimedia and Expo (ICME 2014), 2014-07-14 - 2014-07-18, Chengdu, China.

	Text ICME2014.pdf - ["content_typename_Accepted version (post-print)" not defined] Restricted to Repository staff only Available under License : See the attached licence file. Download (617kB)
	Text (licence) SRI_deposit_agreement.pdf Restricted to Repository staff only Available under License : See the attached licence file. Download (33kB)

Abstract

In this paper, an enhanced perceptual image authentication approach is proposed with extra ability of tamper localization and image self-restoration by combining perceptual hashing and digital watermarking technologies. Compared with other perceptual hashing schemes, this proposed approach could locate the maliciously tampered regions and further recover these regions to some extent. Another advantage of this approach is its robustness to various non-malicious image processing operations. This approach could provide better robustness to most content-based image processing operations such as JPEG compression and additive Gaussian noises than most existing semi-fragile watermarking methods. Experimental results demonstrated the high authentication accuracy rate to non-malicious and malicious image processing operations. Moreover, maliciously tampered regions could be correctly localized and the original images can be recovered with good quality as well.

Item Type:	Conference or Workshop Item (UNSPECIFIED)
Authors :	Name Email ORCID Liu, F Wang, H Cheng, L-M Ho, ATS Li, S
Date :	2014
DOI :	10.1109/ICME.2014.6890128
Contributors :	Contribution Name Email ORCID http://www.loc.gov/loc.terms/relators/PBL IEEE,
Related URLs :	Author Publisher
Depositing User :	Symplectic Elements
Date Deposited :	28 Mar 2017 13:12
Last Modified :	31 Oct 2017 17:00
URI:	http://epubs.surrey.ac.uk/id/eprint/806209

Actions (login required)

View Item

Downloads