Multi-watermarking algorithm for medical image based on NSCT-RDWT-DCT

Abstract

With the rapid development of medical imaging technology, digital information systems have become more and more important. However, medical images are vulnerable to be attacked or malicious tampered during transmission. Medical image watermarking algorithm using wavelet transform can solve the above problems and has been widely used in recent years. But due to its poor resistance to geometric attacks, the security of medical images cannot be guaranteed. In this work, a medical image multi-watermarking algorithm based on NSCT-RDWT-DCT is proposed. First, a more robust feature vector of medical image is extracted by combining non-subsampled contour transform (NSCT), redundant discrete wavelet transform (RDWT) and discrete cosine transform (DCT) to resist geometric attacks. Then, zero watermarking technology is adopted to embed and extract watermarks considering the capacity and invisibility of the medical image. Moreover, Arnold Cat Map is applied to enhance the security of the watermarks information. The experimental results show that the proposed algorithm can effectively extract the watermarked information. Compared with the existing algorithms, it has good performance against both conventional attacks and geometric attacks.