Abstract: Aiming at problems of difficult embedding of Chinese text digital watermarking, weak invisibility and weak robustness, this paper presents a novel Chinese zero-watermarking algorithm based on the number of strokes of Chinese characters. Through combining the frequency of use of Chinese characters in the text, the most frequently used Chinese characters can be got. This paper uses the number of strokes of  the most frequently used Chinese characters as a sequence of the text feature, then generates a set of registration key using this sequence and the watermark information. This experiment result shows that the algorithm achieves simply and has good visibility and robustness.

Key words: text zero-watermarking, frequency of use of Chinese characters, strokes of Chinese characters, robustness

摘要： 针对基于汉字的文本嵌入比较困难，水印不可见性不高，鲁棒性不高等特点，提出一种基于汉字笔画数的文本零水印新算法，算法通过统计各汉字在文本中的使用频率，即可得到使用频率最高的汉字，使用频率最高汉字的笔画数作为文本特征序列，用此序列与水印信息运算生成一组注册码。实验结果表明，该算法实现简单，具有很好的不可见性和鲁棒性。

关键词: 文本零水印, 汉字使用频率, 汉字笔画, 鲁棒性