Research Of Digital Watermark Algorithm Based On DCT Domain

With the development of computer networks and the communication technology, digital work such as image, video, audio may be obtained easily, at the same time care is taken to protect copyright of these work and prevent the pirate. So, digital watermark technique is brought forward to protect the legitimate right of owner of work. On the base of research and discussion of digital watermarking algorithms, the thesis designs and realizes digital watermark algorithm in DCT transform domain, and the robustness is also tested. The fundamentals, such as basic features, principles, classification and the current development status, purpose, significance of digital watermarking are introduced and then the general frame of digital watermarking is presented, which provide theoretical basis for the digital watermark algorithm's proposal, realization and testing. As the significant component of digital watermark algorithms, the designs and realization of meaning watermark and unmeaning watermark are detailedly introduced. What's more, the general features and application in the digital watermark algorithms of pseudorandom real sequence, m-sequence, chaos sequence and error correction code are discussed; spread spectrum technique and basic definition of discrete cosine transform are also presented. Three different digital watermarking algorithms are propounded, which are based on DCT domain. The first algorithm embeds a binary watermark, and the watermark is embedded into all the coefficients of discrete cosine transform of original image. Moreover, the original location of the sorted coefficients can be used for a key, which guarantees the security of watermark. The second algorithm embeds multiple watermarks. The algorithm is based mainly on the theory of spread spectrum technique. It regards the original image as transmission channel and watermark as transmission signal, using different pseudorandom sequences embeds multiple watermark into the same image at the same time and the impact between watermarks is negligible. The third algorithm is the improvement of the above two algorithm. It uses error correction code and sort processing technique, which enhance the robustness and advance detection method. The relation among code length,chaos sequence length and watermark capacity is utilized in order to ascertain the insertion capacity of the third algorithm. The results of computer simulation show that the proposed three watermark algorithms do not lead to the distortion of the original image and they can resist the common image processing operations, such as JPEG compression, filtering and cropping, noise.