Home > Published Issues > 2023 > Volume 14, No. 2, 2023 >
JAIT 2023 Vol.14(2): 242-249
doi: 10.12720/jait.14.2.242-249

A Coloured Image Watermarking Based on Genetic K-Means Clustering Methodology

Zainab Falah Hassan, Farah Al-Shareefi, and Hadeel Qasem Gheni *
Department of Computer Science, Babylon University, Babylon, Iraq
*Correspondence: hajer.s.abbas.uoesraa@gmail.com (H.Q.G.)

Manuscript received June 9, 2022; revised August 10, 2022, accepted October 19, 2022; published March 17, 2023.

Abstract—There are two techniques long-established in image watermarking area, namely the k-means and genetic algorithms. The first one is commonly used to allocate an image’s pixels into distinct clusters. However, the allocation of these pixels is not optimal in all cases. The second technique is usually employed to produce an optimal watermarking solution. In this paper, a hybrid methodology is presented for coloured image watermarking that integrates both genetic algorithm and k-means clustering activity to attain the optimized cluster centroids. These centroids are utilized to optimally distribute the pixels of the cover and watermark images into suitable clusters. This will help decrease the perceptible changes in the watermarked image with the naked eye. For concealment, the Least Significant Bits method is adopted. Typically, the pixels of every watermark cluster are concealed in its closest cover’s cluster; wherein every two successive pixels hide the bits of a single cover image’s pixel. The experimental results demonstrate that the proposed methodology satisfies a sufficient imperceptibility that yields and boosts resistance against common attacks.
Keywords—genetic algorithm, LSB method, K-means clustering, watermarking techniques, attack types, performance measures

Cite: Zainab Falah Hassan, Farah Al-Shareefi, and Hadeel Qasem Gheni, "A Coloured Image Watermarking Based on Genetic K-Means Clustering Methodology," Journal of Advances in Information Technology, Vol. 14, No. 2, pp. 153-159, 2023.

Copyright © 2023 by the authors. This is an open access article distributed under the Creative Commons Attribution License (CC BY-NC-ND 4.0), which permits use, distribution and reproduction in any medium, provided that the article is properly cited, the use is non-commercial and no modifications or adaptations are made.