Research On Watermarking Algorithm For Color Images In The Wavelet Transform Domain

With the development of society and the progress of science and technology, digitization has become an important means of information, but without distortion and unlimited copies of the natural characteristics of digital technology has caused severe threat to the lawful rights and interests of the digital product owner, the copyright protection of digital products become an urgent need to solve. Digital watermark can provide effective protection for the copyright of digital products, and increasingly become the research hotspot of copyright protection areas.At present, the image watermarking algorithm in wavelet domain is one of the digital watermarking algorithms in transform domain, and is widely used. But the carrier images in most of these algorithms are gray image, and the watermark images are usually binary image, and the watermarking algorithm that is specifically for color image is very few. After studying the theory of digital watermarking and the human visual system, this paper puts forward two digital watermarking algorithms in DWT domain, and the research objects of the algorithms are color images.The first one is an adaptive image watermarking algorithm in DWT domain based on HVS. This algorithm splits the low-and-middle coefficients which are got from the DWT of carrier images into several sub-blocks of the same size, chooses the blocks which have larger variance to embed the watermark according to HVS, and realizes the adaptive embedding strength.The second one is a color image watermarking algorithm in DWT domain based on DCT and SVD. This algorithm combines discrete cosine transform, discrete wavelet transform and singular value decomposition, and takes advantage of them. The core ideology is that the DCT coefficients of the watermarking image are embedded into the SVD of low-and-middle frequency components with new method.The experimental results show that both of the two algorithms have good robustness and invisibility, and have a good performance on resistance to the common attacks. The first algorithm has better invisibility than the second one; the second algorithm has better robustness and larger embedding capacity than the first one, and is valuable to the study of color image watermarking algorithm. The two algorithms also reflect the advantage of DWT in the digital watermarking algorithms.