¡@

A lossless large-volume data hiding method based on histogram shifting is proposed. The method is based on a scheme of hierarchically dividing a cover image into smaller blocks for data embedding using the histogram shifting technique, which yields a large data hiding capacity and results in a high stego-image quality. A technique for recursive looking-ahead estimation of the maximum data hiding volume at the lowest level of the block-division tree structure is proposed to yield an optimal data hiding result. The technique is shown to break a bottleneck of data-hiding-rate increasing at the image block size of 8 X 8, which is found to exist in other histogram-shifting methods. Four ways of block divisions are used, and one of them is selected optimally in each tree level of block divisions. A series of experiments have been conducted, and superiority of the proposed method to other related methods is shown by comparing the respective experimental results. A good property of the proposed method observed in the experiments is also pointed out, that is, the data hiding rate yielded by the proposed method increases without degrading the stego-image quality, which is an unusual phenomenon in data hiding researches.

Keywords: lossless data hiding, histogram shifting, block division, stego-image, steganography

Retrieve PDF document (201107_05.pdf)

Received March 22, 2010; revised May 27 & July 20, 2010; accepted August 26, 2010.
Communicated by Chung-Lin Huang.
^* This work was supported by the National Science Council of Taiwan, R.O.C., project No. 97-2631-H-009-001.