Directly embedding the secret data into the VQ-compressed domain is practical for
reducing storage and transmittal bandwidth. Many steganography techniques are currently
exploited for the VQ index table. However, embedding strategies extend the amount of
compression needed for embedding secret data and distort the original quality of the VQ
decompressed image. To satisfy the essentials of increasing the secret payload, preserve the
VQ index table without loss, and reduce the compression rate, the proposed scheme utilizes
search order coding to find the correlation between the VQ indices and neighboring indices.
Due to the similarity of adjacent indices, the new scheme can effectively embed the secret
stream into the index table and encode the embedded index table with fewer bits. The
experimental results indicate that the new scheme can achieve superior outcomes of the
compression rate and embedding efficiency than related steganography techniques.
Relation:
International Journal of Signal Processing, Image Processing and Pattern Recognition 2(2):39-51
