Please use this identifier to cite or link to this item:
http://hdl.handle.net/2080/2375
Title: | On Parameterization of Block based Copy-Move Forgery Detection Techniques |
Authors: | Wadhwa, J Ahemad, T Naskar, R Dixit, R |
Keywords: | Copy-move forgery Cyber-crime Digital image forgery False negative rate False negative rate Parameterization |
Issue Date: | Oct-2015 |
Publisher: | ACM |
Citation: | Research in Adaptive and Convergent Systems (RACS 2015), Prague, Czech Republic, 09-12 October 2015 |
Abstract: | With high increase in cyber-crime along with continuous development in multimedia processing and editing technologies, the credibility of digital images is highly at stake in the present day. Digital images act as the major source of legal evidence in various domains such as media, broadcast and legal industries. Hence any form of illegal modifications to them is intolerable. Recently a lot of researchers have focused on detection and control of digital image forgeries. However the literature lacks a standard way of evaluating and comparing the efficiencies of diverse forgery detection techniques. In this paper we propose a standard platform for estimating the efficiencies of state-of-the-art digital image forgery detection techniques. In this work we have dealt with a specific class of digital image forgery: the Copy-Move forgery, which is one of most prevalent forms of attack on digital images. We have compared and analyzed different copy-move forgery detection techniques using the proposed parameters. Our results prove the efficiency of the proposed parameterization and help to select the most suitable scheme according to the user’s requirements. |
Description: | Copyright belong to proceeding publisher |
URI: | http://hdl.handle.net/2080/2375 |
Appears in Collections: | Conference Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
OnParameterization_ACM_2015.pdf | 2.29 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.