Scholarlink

In this work, lifting wavelet transform (LWT)-based medical image watermarking (MIW) utilizing QR decomposition and singular value decomposition (SVD) is proposed in YCbCr color space. Two-level LWT is performed on Y component of host image to divide it into several sub-bands. HH sub-band is further utilized to process with QR decomposition and SVD. Before embedding, watermark is processed with 1-level LWT and SVD. Singular values (SVs) of the watermark and strength factor are added to the SVs of the host image to modify it. To evaluate the imperceptibility and robustness of proposed technique, parameters such as peak signal-to-noise ratio (PSNR), bit error rate (BER), normalized correlation coefficient (NC), and structural similarity index measure (SSIM) are utilized. Robustness is assessed by applying various attacks. Comparison of proposed scheme performance with some prevailing techniques is also presented.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

QR Decomposition and SVD-Based Medical Image Watermarking Technique in Transform Domain

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Ranjana Dwivedi,
  • Divyanshu Awasthi,
  • Vinay Kumar Srivastava

摘要

In this work, lifting wavelet transform (LWT)-based medical image watermarking (MIW) utilizing QR decomposition and singular value decomposition (SVD) is proposed in YCbCr color space. Two-level LWT is performed on Y component of host image to divide it into several sub-bands. HH sub-band is further utilized to process with QR decomposition and SVD. Before embedding, watermark is processed with 1-level LWT and SVD. Singular values (SVs) of the watermark and strength factor are added to the SVs of the host image to modify it. To evaluate the imperceptibility and robustness of proposed technique, parameters such as peak signal-to-noise ratio (PSNR), bit error rate (BER), normalized correlation coefficient (NC), and structural similarity index measure (SSIM) are utilized. Robustness is assessed by applying various attacks. Comparison of proposed scheme performance with some prevailing techniques is also presented.