Authors : K.R. Prabha and M. Jagadeeswari
Abstract: A novel reversible data hiding and compression scheme to embed high capacity of secret bits into a Vector Quantization (VQ) and Side Match Vector Quantization (SMVQ) based compressed image and recover cover image after data extraction is proposed. This study, uses Enhanced Imperialist Competitive Algorithm (EICA) that uses contrast sensitivity as the fitness function to determine threshold value to signify embedding rate for each region of the cover image based on the size of the secret message for optimal embedding capacity and to achieve exact recovery of cover image. The output size of code stream is preserved hiding two secret bits into a single index value during data hiding. For complete recovery of the cover image after the secret extraction and to achieve a high compression ratio Discrete Cosine Transform (DCT) transform and Burrows Wheeler Transform (BWT) transform is used before the quantization. DCT provides excellent energy compaction and lossless BWT reorders the symbols according to their following context. Thus, the proposed method embeds secret bits and compresses the image with secret data and gives the output as code streams with preserved size. Experimental results show that the proposed scheme achieves high embedding capacity and compression rate.
K.R. Prabha and M. Jagadeeswari, 2017. VQ, SMVQ Based Efficient Reversible Image Data Hiding with EICA Method. Asian Journal of Information Technology, 16: 399-409.