Open AccessReversible Watermarking Technique for Fingerprint authentication based on DCT
Author(s) -
Mohamed Lebcir,
Suryanti Awang,
Ali Benziane
Publication year - 2020
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/769/1/012070
Subject(s) - minutiae , watermark , digital watermarking , discrete cosine transform , fingerprint (computing) , artificial intelligence , computer science , computer vision , fingerprint recognition , authentication (law) , pattern recognition (psychology) , image (mathematics) , mathematics , computer security
In this paper, a new reversible and blind fingerprint image watermarking technique based on the differential method and discrete cosine transform (DCT) domains is presented. The focus is to increase the security of the fingerprint image in authentication systems. Two DCT-transformed sub-vectors are employed to embed the bits of the watermark sequence in a differential scheme. The original sub-vectors are acquired by the DCT transform on the host fingerprint image. In the extraction process, a minor variance between the sub-vectors that correspond to the watermarked fingerprint image directly allows access to the embedded watermark sequence; therefore, the extraction process doesn’t require an original fingerprint. The original fingerprint image is then recovered from the watermarked fingerprint image based on the reversible watermarking technique. The similarity between the reversible fingerprint image and the original is considered, and we could extract minutiae points from it without a problem. The proposed technique is evaluated using 80 fingerprint images from 10 persons for each from FVC2002 fingerprint database. Eight fingerprint images have been taken from each person to be used as the template then the process was followed by embedding the watermark into each fingerprint image. The experimental results validate the proposed method able to give promising results in preserving the fingerprint image security.