A reversible data hiding scheme based on maximum histogram gap of image blocks

In this paper a reversible data hiding scheme based on shifting the histogram of host image blocks is presented. This method attempts to use full available capacity for data embedding by dividing the image into non-overlapping blocks. Applying histogram shifting to each block requires that extra information to be saved as overhead data for each block. This extra information (overhead or bookkeeping information) are used in order to extract payload and recover the block to its original state. A method to eliminate the need for this extra information in this paper is presented. This method uses maximum gap that exists between histogram bins for finding the value of pixels that was used for embedding in sender side. Experimental results show that while this method has higher embedding capacity than the original histogram method and their improved schemes, it maintains the quality of watermark image at an acceptable level.