Fuzzy Based Image Forensic Tool for Detection and Classification of Image Cloning

DOI

10.1080/18756891.2016.1161364How to use a DOI?

Keywords

Image forensics; Cloning detection; Discrete Wavelet Transform (DWT); Discrete Cosine Transform (DCT); Singular Value Decomposition (SVD); Principal Component Analysis (PCA)

Abstract

With the easy availability of image processing and image editing tools, the cases of forgery have been raised in the last few years. Now days it is very difficult for a viewer and judicial authorities to verify authenticate a digital image. Cloning or copy-move technique is widely used as forgery to conceal the desired object. To hide various type of forgery like Splicing (compositing), cloning (copy-move) etc., various post-forgery techniques like blurring, intensity variation, noise addition etc. are applied. To overcome the mentioned difficulty, a forgery detection tool must comprise of several detection algorithms which work collaboratively to detect all the possible alterations and provide a single decision. This paper presents a universal tool comprising PCA, DWT, DWT-DCT, DWT-DCT-SVD, DFT, DCT, DWT-DCT (QCD) techniques used for reduction, feature vector calculation and thus detecting forgery. Due to varied, erroneous, heterogeneous output of different reduction methods, it is very difficult to recognize the pre-processing done with available various classification systems. A fuzzy inference system has been developed to authenticate, find extend of forgery, parameters of forged area, robustness and accuracy of all the 7 detection tools, and the type of processing done on tempered image. Experimental results have shown that our classification system achieves accuracy of 94.12% as regards subjection to transformations like Blurring, Intensity Variation and Gaussian Noise Addition, JPEG compression, normal forgery (other random transformations). Two different membership functions are taken in this fuzzy system and different if-then rules are defined for classification of different types of pre-processing performed on the image.

Copyright

© 2016. the authors. Co-published by Atlantis Press and Taylor & Francis

Open Access

This is an open access article under the CC BY-NC license (http://creativecommons.org/licences/by-nc/4.0/).

Download article (PDF)
View full text (HTML)

TY  - JOUR
AU  - Mohammad Farukh Hashmi
AU  - Avinash G. Keskar
AU  - Vikas Yadav
PY  - 2016
DA  - 2016/04/01
TI  - Fuzzy Based Image Forensic Tool for Detection and Classification of Image Cloning
JO  - International Journal of Computational Intelligence Systems
SP  - 351
EP  - 375
VL  - 9
IS  - 2
SN  - 1875-6883
UR  - https://doi.org/10.1080/18756891.2016.1161364
DO  - 10.1080/18756891.2016.1161364
ID  - Hashmi2016
ER  -