Copy-Move Forgery Detection Using DyWT

View Sample PDF

Author(s): Choudhary Shyam Prakash (Indian Institute of Technology, Department of Computer Science and Engineering, Dhanbad, India)and Sushila Maheshkar (Indian Institute of Technology, Department of Computer Science and Engineering, Dhanbad, India)
Copyright: 2020
Pages: 10
Source title: Digital Forensics and Forensic Investigations: Breakthroughs in Research and Practice
Source Author(s)/Editor(s): Information Resources Management Association (USA)
DOI: 10.4018/978-1-7998-3025-2.ch010

Keywords: Criminal Science and Forensics / Digital Crime & Forensics / Information Science Reference / Security & Forensics

Purchase

View Copy-Move Forgery Detection Using DyWT on the publisher's website for pricing and purchasing information.

Abstract

In this paper, we proposed a passive method for copy-move region duplication detection using dyadic wavelet transform (DyWT). DyWT is better than discrete wavelet transform (DWT) for data analysis as it is shift invariant. Initially we decompose the input image into approximation (LL1) and detail (HH1) sub-bands. Then LL1 and HH1 sub-bands are divided into overlapping sub blocks and find the similarity between the blocks. In LL1 sub-band the copied and moved blocks have high similarity rate than the HH1 sub-band, this is just because, there is noise inconsistency in the moved blocks. Then we sort the LL1 sub-band blocks pair based on high similarity and in HH1 blocks are sorted based on high dissimilarity. Then we apply threshold to get the copied moved blocks. Here we also applied some post processing operations to check the robustness of our method and we get the satisfactory results to validate the copy move forgery detection.

The IRMA Community

Research IRM

Copy-Move Forgery Detection Using DyWT

Purchase

Abstract

Related Content

IRMA Sponsors