Context-adaptive binary arithmetic coding

Mathematics

Data compression

Biology

Algorithm

Image (mathematics)

Statistics

Genus

Synonym (taxonomy)

Digital watermarking

Coding (social sciences)

Arithmetic coding

Artificial intelligence

Watermark

Programming language

Substitution (logic)

Computer science

Botany

Binary number

Decoding methods

Arithmetic

Cmc-computers Materials & Continua

Abstract: For protecting the copyright of a text and recovering its original content harmlessly, this paper proposes a novel reversible natural language watermarking method by combining arithmetic coding and synonym substitution operations. By analyzing relative frequencies of synonymous words, synonyms employed for carrying payload are quantized into an unbalanced binary sequence, which is redundant. The quantized binary sequence is compressed by arithmetic coding losslessly to provide a spare for accommodating additional data. Then, the compressed data appended with the watermark are embedded into the cover text via synonym substitutions in an invertible manner. On the receiver side, the watermark and compressed data can be extracted by decoding the values of synonyms in the watermarked text, so that the original context can be perfectly recovered by employing arithmetic coding to decompress the extracted compressed data and substituting the replaced synonyms with their original synonyms. Experimental results illustrate that the proposed method can extract the watermark successfully and recover the original text losslessly. Additionally, it achieves a high embedding capacity.

Reversible Natural Language Watermarking Using Synonym Substitution and Arithmetic Coding