A Watermarking Algorithm for Gray-Level Watermark Based on Local Feature Region and SVD

In this paper, a novel gray-level image watermarking algorithm based on local feature region and Singular Value Decomposition (SVD) is presented for copyright protection. A new way which utilizes the Binary Coded Decimal (BCD) method and encodes the gray-level secret information into bit string is used. It employs the singular values of SVD to extract the most valuable features of the watermark image and reduces lots of redundant information. The feature regions of the host image, which are used for information hiding, can be correctly detected by the improved Harris-Laplace corner detector even after signal processing and geometric attacks. The application of the proposed scheme makes the gray-level watermark capable of transparent hiding. Simulation results show that our method performs better than traditional methods under different attacks.