A reversible data hiding scheme for VQ indices using histogram shifting of prediction errors

Concomitant with the rapid advancements in information technology, the issue of secure data transmission through the Internet has become increasingly important. Hiding data in images is an important technique in digital media; it facilitates confidential data transfer to receivers and renders the data virtually undetectable by third parties. In this paper, a novel reversible data hiding scheme based on vector quantization (VQ) is proposed. First, codebook sorting is employed to enhance the correlation of neighbor indices and then a prediction error technique is used to generate a high peak histogram. The secret data are then embedded via histogram shifting of prediction errors. Our proposed scheme utilizes only one codebook, unlike other similar methods, and the stego carrier (index table after embedding) can be decompressed without secret data extraction to avoid detection by third parties. Experimental results indicate that the proposed scheme has better visual image quality and greater embedding capacity than recently proposed schemes of a similar nature.