<p>Dual complex matrix decomposition plays an important role in color image watermarking due to its effectiveness in preserving structural and chromatic information. In this paper, we utilize the relevant theoretical knowledge of dual complex number and dual complex matrix to propose the dual complex Householder transformation. Using the dual complex Householder transformation, we propose the dual complex Hessenberg decomposition algorithm, which is more efficient than quaternion Hessenberg decomposition algorithm. Based on dual complex Hessenberg decomposition, we propose a novel blind color watermark scheme where the watermark embedding process is designed to transform the encrypted binary watermark information into the embedded information to replace the low-energy elements in the upper Hessenberg matrix. Experimental results indicate the proposed scheme has strong invisibility and robustness compared with other related schemes.</p>

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

Dual complex Hessenberg decomposition method for blind color image watermarking

  • Tao Wang,
  • Ying Li,
  • Mingcui Zhang,
  • Wenxv Ding,
  • Musheng Wei

摘要

Dual complex matrix decomposition plays an important role in color image watermarking due to its effectiveness in preserving structural and chromatic information. In this paper, we utilize the relevant theoretical knowledge of dual complex number and dual complex matrix to propose the dual complex Householder transformation. Using the dual complex Householder transformation, we propose the dual complex Hessenberg decomposition algorithm, which is more efficient than quaternion Hessenberg decomposition algorithm. Based on dual complex Hessenberg decomposition, we propose a novel blind color watermark scheme where the watermark embedding process is designed to transform the encrypted binary watermark information into the embedded information to replace the low-energy elements in the upper Hessenberg matrix. Experimental results indicate the proposed scheme has strong invisibility and robustness compared with other related schemes.