A blind and robust color image watermarking scheme based on DCT and DWT domains

• Published in: Multimedia tools and applications 2023-09, Vol.82 (21), p.32855-32881
• Main Authors: Mohammed, Abdulhakeem O., Hussein, Haval I., Mstafa, Ramadhan J., Abdulazeez, Adnan M.
• Format: Article
• Language: English
• Subjects: Artificial intelligence; Color imagery; Computer Communication Networks; Computer Science; Data Structures and Information Theory; Decomposition; Digital data; Discrete cosine transform; Discrete Wavelet Transform; Embedding; Internet of Things; Multimedia Information Systems; Special Purpose and Application-Based Systems; Theft; Watermarking; Wavelet transforms
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-023-14797-0

With the emergence of the Internet of Things (IoT) and many smart gadgets that support artificial intelligence, it is easier than ever to acquire, reproduce, and disseminate a large number of digital data. However, these great technologies have made it possible for intruders to easily violate issues related to copyright protection, identity theft, and privacy leakage. To address such issues, several approaches have been developed, among which image watermarking has been proven to be an ideal solution. In this paper, a blind watermarking approach for RGB color images based on joint Discrete Cosine Transform (DCT) and Discrete Wavelet Transform (DWT) is proposed. First, a self-adaptive color selecting strategy is used to select either the blue or green channel of the host image for embedding purpose. Subsequently, the selected color is subdivided into non-overlapping square blocks of size 4 × 4, and then DCT is applied to each block. Afterward, the DC values obtained from each block are decomposed into four sub-bands using DWT, and then the LH middle frequency sub-band is further decomposed into four sub-bands using DWT. Lastly, the LH1 obtained from LH is utilized for watermark embedding. To provide security to the proposed approach, the watermark image is encrypted before embedding using a chaotic sequence originated from a logistic map method. Experimental results reveal that the proposed approach not only enhances watermark invisibility but also provide excellent watermark robustness, meeting the main requirements of image watermarking.