Research On Watermarking For Two-dimensional Vector Map Based On Transform Domain

With the rapid development of the Internet technology, the dissemination of information is becoming easier and faster, which lead to the copyright protection and the legality certification of digital resource of information become hot spots in security technology in the network environment. For their advantages such as high precision, small data size, easiness to operation, lossless scaling and high cost, digital watermark seems to be an effective method to be used in data security maintenance, which provides a new way for vector digital map copyright protection.This paper focuses on the watermarking techniques for two-dimensional (2D) vector map. Main works of this thesis are as follows:First, expounds the principle, character, classification and capacity value of digital watermarking technique, analyzes current research status of digital watermarking technique for 2D vector map, and summarizes the shortage of actual watermarking technique. After that, we introduce the relevant theoretical knowledge, including chaos theory, discrete cosine transform and flourier transform.Secondly, a 2D digital watermarking algorithm based on chaos scrambling and Discrete Cosine transformation is proposed. By using a designed formula of watermark intensity, the algorithm embeds scrambled watermarking information into low-frequency coefficients of the blocks of 2D vector, which was transformed by discrete cosine transform, and it strengths robustness of the watermark.Thirdly, a blind watermarking algorithm of 2D vector map based on chaos scrambling and discrete wavelet transform is put forwarded. Considering the shortages of some watermarking algorithms wavelet domain such as large coordinates deviation and small watermark capacity, we use chaos scramble the watermark information, extract the original sequence for vector map vertex order, then 2-rank discrete wavelet transform is done to them, consequently embed the watermark into the middle frequency coefficients, while the embedded data is depended on the relationship between the wavelet coefficients and the embedded watermark..Finally, experiment and simulation results prove that the algorithm can guarantee the precision of vector map data, and can effectively resist attacks such as translation, scaling, revolving, shear and etc. The robustness of the algorithm is greatly improved.