TY  - MANSCPT
AU  - Siti Nur Avivah,
TI  - A robust watermark embedding scheme using non-subsampled shearlet transform and lifting wavelet transform for digital images
SN  - THE0010074 (Local)
PY  - 2025///
CY  - Kuantan, Pahang
PB  - UMPSA
KW  - Faculty of Computing
KW  - Dissertations
KW  - Universities and colleges
KW  - Theses
N1  - Faculty of Computing; Thesis (Master of Science) -- Universiti Malaysia Pahang – 2025; Includes bibliographical references
N2  - Digital images are one of the most widely utilized media due to their capacity to transmit  information visually. The rapid advancement of network technologies and social media  over recent decades has facilitated the widespread distribution of images via the internet.  However, this ease of distribution allows unauthorized users to copy and modify the  image, potentially infringing on copyright. Digital image watermarking is a robust  method to combat copyright violations of digital content. Watermarking can be performed using either single or multiple watermarks. This study proposes multiple  watermarking schemes based on Non-Subsampled Shearlet Transform (NSST) and  Lifting Wavelet Transform (LWT) using a small block of 4×4. The watermark is  embedded specifically in the red and green channels of the host images. NSST-LWT  transforms the image channels. Then, the entropy and edge entropy of the LL frequency  is calculated to determine the suitable blocks for embedding the watermark based on  Human Visual System (HVS). The selected blocks are then decomposed by using DCT.  The proposed scheme identifies the optimal embedding locations within the 4×4 DCT  coefficient matrix. The selected coefficients at (1,1), (1,2), (2,1), and (2,2) are then  transformed using SVD. The singular value in a position of (1,1) is then used to  incorporate the watermark bits with a certain rule. The selected DCT coefficients  influence the quality and strength of the watermarked image. The experimental results  reveal that the suggested scheme has strong imperceptibility and resilience under various  manipulation attacks, with average PSNR and SSIM values of 46.88 dB and 0.9981,  respectively. Additionally, the robustness performance regarding correlation (NC) values  is close to 1
ER  -