High-capacity blind binary text document watermarking scheme robust to print-and-scan operations

Abstract

An algorithm of high-capacity blind text watermarking is presented that is robust to print-and-scan operations. First, binary images of Chinese text characters are segmented depends on the complexity of the characters: each character image is segmented to embedded part, adjust part, and dislodge part. After segmentation, a quantization function is built on the basis of invariables during printing-scanning. The watermark is embedded via the strategy of flipping pixels at the characters boundaries. This method exploits human visual masking characteristics to reduce the distortion degree in the text image. Experimental results indicate that the method is robust to common attacks and the watermark can be extracted without original document. © 2012 Springer-Verlag.