Title: Augmenting the detection of online image theft using diagonal transformation and iris biometric watermarking

Authors: Jincy J. Fernandez; Nithyanandam Pandian; Raghuvamsh Chavali; Ashwanth Kumar Appalaghe

Addresses: VIT University, Chennai, India ' School of Computing Science and Engineering, VIT University, Chennai, India ' Rashonic Unicpixel Pvt. Ltd., Telangana, India ' Rashonic Unicpixel Pvt. Ltd., Telangana, India

Abstract: The digitisation of data has brought in various social media platforms where a lot of data are compromised every day. Likewise, there is a possibility of theft for every photo published online. Hence, the research work proposes an image theft detection technique where biometric traits of the owner of the photo, are embedded into the image. This helps to prove rightful ownership in case of any ownership disputes on the image. The proposed architecture identifies the rightful owner by comparing the embedded traits from the claimed image with that of the person asserting ownership. The watermark imperceptibility is analysed and robustness against various attacks is proven. Also, the accuracy of the work is analysed in terms of the false acceptance rate and false rejection rate. By considering iris images taken from CASIA, and IIT-Delhi databases, the proposed approach achieved a better accuracy rate with a very low FAR and FRR.

Keywords: biometric watermarking; pixel adjacency; matrix embedding; iris recognition.

DOI: 10.1504/IJICS.2022.127172

International Journal of Information and Computer Security, 2022 Vol.19 No.3/4, pp.321 - 345

Received: 25 Dec 2020
Accepted: 12 Mar 2021
Published online: 23 Nov 2022 *

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