Title: Multi-image authentication in frequency domain

Authors: Anirban Goswami; Nabin Ghoshal

Addresses: Department of Information Technology, Techno India, EM 4/1, Salt Lake, Sec-V, Kolkata – 700091, West Bengal, India ' Department of Engineering and Technological Studies, University of Kalyani, Kalyani, Nadia-741235, West Bengal, India

Abstract: Today multimedia documents are widely used for data authentication. In this paper, we propose a model for image authentication with robustness against certain attacks. To achieve this, we have effectively utilised two distinct colour cover images simultaneously and applied discrete cosine transform to convert them from spatial to frequency domain. The payload data, payload size and a 160 bit message digest are inserted in blocks of pixels of size 2 × 2 taking alternately from the two images. Capacity of embedding is increased by fabricating two payload bits in each converted frequency component whereas for enhanced security we have used pseudo-random insertion positions for the payload bits. On the receiving end the integrity of the extracted payload is verified by using the message digest. The algorithm has been tested effectively against certain steganographic attacks. Also experimental results have been analysed by using different metrics and compared with other similar algorithms.

Keywords: image authentication; discrete cosine transform; DCT; inverse discrete cosine transform; IDCT; message digest; MD; statistical attack; visual attack; collusion attack; HSI.

DOI: 10.1504/IJICT.2017.086249

International Journal of Information and Communication Technology, 2017 Vol.11 No.2, pp.166 - 186

Received: 05 Jul 2014
Accepted: 29 Oct 2014
Published online: 04 Sep 2017 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article