Image forgery detection in contourlet transform domain based on new chaotic cellular automata

Abstract

Image authentication includes techniques that detect unauthorized changes in digital pictures. Image authentication is generally categorized into two primary categories, active and passive methods. In this paper, a new active image authentication method is proposed for tamper detection based on contourlet transform. Cellular automata and ‘Game of Life’ are also used to improve the detection of tampered area. A pseudo‐random number generator based on cellular automata has been used to improve security of proposed technique. We evaluates security and randomness of proposed pseudo‐random number generator using by NIST statistical test suite. Experimental results show that the proposed method is sensitive to any tampers, and it can detect forged areas of the images. Visual quality of image after embedding process is desirable due to Peak Signal‐to‐Noise Ratio (PSNR) measure. Copyright © 2015 John Wiley & Sons, Ltd.

Taylor-rider-based deep convolutional neural network for image forgery detection in 3D lighting environment

“…Knowing the importance of image forensic techniques, researchers have contributed a lot in the same field for the last few decades (Spagnolo and De Santis, 2011). To deal with security weaknesses and adding the concept of cryptography in the production of the authentication sequence, a 3D quantum chaos map (Barani et al, 2015(Barani et al, , 2019 has been used to boost the algorithm key length. The weakness of digital watermarking (Cox et al, 2002;Katzenbeisser and Petitcolas, 2000) is that a watermark must be introduced at the time of recording, which limits this technique to specifically equipped digital cameras (Farid, 2009).…”