Research On The Print-Scan Resistant Watermarking Technology And Application

The print-scan resistant watermarking technology is that the watermarking embedded in the digital image can be extracted correctly when the image is reproduced by print and scan operation. The impact to the image given by the process of print and scan is complex, because of the two times conversion of types from D/A to A/D, and the manual operation miss. Print-scan attack summarized as grayscale conversion attack, geometric attack and lowpass filter attack is a challenge for most of digital watermarks. The print-scan resistant watermarking technology can be used in several fields such as copyright protection, origin certification and preventing forgery for printed publication.In this thesis, we focus on the research of the print-scan resistant watermarking algorithms, and also design and implement the robustness watermarking system which can resist the print-and-scan. The content of the thesis mainly includes:Print-scan resistant watermarking algorithm based on the DFT; The text watermarking algorithm resistant to print-scan; The design and implement of the leak-tracking management system for the paper document. The specific research contents and research results are as follows:1. Based on the design framework of print-scan resistant watermarking algorithm, we propose a algorithm which combines the methods of image complexity and discrete Fourier transform (DFT). We determine which block can be used to embed watermark by computing the local image complexity of the block, considering the masking of the human visual system. The red component of the block is transformed with DFT and one bit is embedded in the amplitude of DFT coefficients. The hidden data can be extracted without using the original image to verify the copyright of the reproduced image. Experimental results suggest that the proposed method is robust after high-quality printing and scanning.2. A high-capacity text watermarking scheme resistant to print-and-scan processes is proposed. The binary document image is first segmented into individual characters. A multi-bit watermark is then embedded by flipping the boundary pixels of each character based on a print-scan invariant quantity and complexity of the character. A quantitative model is established to achieve the character pixel quantization, which can minimize visual distortion. Experimental results show that the proposed scheme can increase embedding capacity. It can effectively resist print-scan attacks so that watermark hardcopy documents can be put into practice.3. A printed text authentication algorithm is proposed based on the mathematical expression of the Chinese character which is used to describe the component of the Chinese character. Zero-watermark as fundation for the copyright is derived by the mathematical expression of the selected Chinese character randomly and the operators of whole text. The final judgement is determined by the comparison between the extracted bits and zero-watermark bits. The zero-watermark technology guarantee no distortion for no change to the text. This algorithm can be used for the copyright protection of printed publication because it resist to pint-and-scan.4. A paper document management system to detect information leaks is implemented, Besides the watermark embedding and extracting, we design the error-control, denosing and document skew detection for the utility system, which guarantee the stability and veracity. The system embeds the invisible watermark bearing personal identity information into the paper document through virtual printing technology. It can be used to detect the source of leaks by extracting the print-scan resistant watermarking. The performance of system has reached the practical requirements by the real environment testing.