Abstract
In this paper, we present a block based reversible data hiding algorithm using the edge direction scan order. In the proposed algorithm, we first divide an image into a set of blocks. Then, we develop a novel feature prediction scheme to classify the blocks into non-smooth and smooth blocks as well as predict the edge directions of the blocks. Based on the four kinds of edge directions, four edge direction scan orders of blocks are presented to gain a better prediction as well as to guide the embedding process with a higher embedding capacity. Finally, we exploit the difference expansion conducted by the proposed edge direction scan order to embed hidden data. In addition, since the appropriate threshold determination is a critical issue for a data hiding algorithm, we design a new systematic way to tackle this problem. The results of experiments on some typical test images demonstrate that, under the same embedding capacity, the proposed reversible data hiding algorithm delivers better quality of marked images than the three previously published reversible data hiding algorithms. Conversely, under the same quality of marked images, the proposed algorithm has higher peak signal/noise ratio performance.