New Framework for Code-Mapping-based Reversible Data Hiding in JPEG Images

Code mapping (CM) is an efficient technique of reversible data hiding (RDH) in JPEG images, which embeds data by constructing the mapping relationship between used codes and unused codes in JPEG bitstream. In this paper, we present a new framework to design the CM-based RDH method. Firstly, to suppress the file size expansion and improve the applicability, a new code mapping strategy is proposed. Based on the proposed strategy, the mapped codes are redefined by customizing a new Huffman table thoroughly rather than selected from the unused codes in the original Huffman table. Afterwards, the key issue of designing the CM-based RDH method, i.e., constructing the code mapping, is converted into solving a combinatorial optimization problem. As a realization, a novel CM-based RDH method is introduced by employing the genetic algorithm (GA). Experimental results show that the efficacy of the proposed method with high embedding capacity and no signal distortion while suppressing file size expansion.