Please use this identifier to cite or link to this item: https://doi.org/10.1109/CGIV.2008.13
Title: Tampered image detection using image matching
Authors: Li, Z.
Nee, A.Y.C. 
Ong, S.K. 
Gong, W.
Issue Date: 2008
Source: Li, Z., Nee, A.Y.C., Ong, S.K., Gong, W. (2008). Tampered image detection using image matching. Proceedings - Computer Graphics, Imaging and Visualisation, Modern Techniques and Applications, CGIV : 174-179. ScholarBank@NUS Repository. https://doi.org/10.1109/CGIV.2008.13
Abstract: With the development of image processing software, such as Photoshop, scientific analysis for detecting tampered images becomes a critical research issue. A hybrid image matching algorithm for image analysis is proposed and evaluated. Features are extracted using blob detectors and interest point detectors. This combination ensures sufficient correspondences in small image patches while maintaining high accuracy. Nearest neighbors are searched using hybrid spill tree in order to reduce the computational load. Applying the algorithm to tampered image detection, a novel solution is proposed utilizing the scale space information and gradient orientation information from the corresponding features. Furthermore, an order regularity of scale space between the corresponding patches and a description for object rotation are also proposed. In addition, we conducted experiments on tampered images, proving the method to be powerful. © 2008 IEEE.
Source Title: Proceedings - Computer Graphics, Imaging and Visualisation, Modern Techniques and Applications, CGIV
URI: http://scholarbank.nus.edu.sg/handle/10635/73915
ISBN: 0769533590
DOI: 10.1109/CGIV.2008.13
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
There are no files associated with this item.

SCOPUSTM   
Citations

4
checked on Dec 14, 2017

Page view(s)

16
checked on Dec 10, 2017

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.