Text Coverless Information Hiding Based on the Combination of Chinese Character Components

In recent research works on coverless text information hiding, there are two main problems to be solved: low hiding success rate and small hiding capacity caused by the small amount of redundant information in the cover text. In order to overcome these problems, a method of coverless information hiding based on the combination of Chinese character components is proposed. This method is based on the traditional "location tag + keywords" mode for coverless information hiding, but with an improved design to extend the scope of available keywords for data hiding beyond the original cover texts. In our proposed method, each Chinese character in the keyword is split into two components: radicals and independent Chinese character, which are collected and will be recombined in pairs to generate the new Chinese characters. This method increases the hiding success rate and hiding capacity by generating Chinese characters that do not exist in the original cover text. The experimental result shows that the proposed method can utilize less size of carrier text database for efficient information hiding. When the size of carrier text database is reduced to 198MB, the highest hidden capacity and hidden success that can be achieved are 13.45 and 0.99, respectively.