Adaptive reversible data hiding scheme based on prediction error histogram shifting to achieve high embedding capacity and keep image visual quality

Abstract

Today, with the development of processing technologies and improvement in cloud networks, a large number of photos are transferred in networks. The need to hide data and maintain security attracted the attention of researchers. In this paper, a reversible data hiding method based on the prediction error histogram shifting is proposed in order to increase the embedding capacity and maintains the visual quality of the image as well. To reach these goals, the original image is transformed into block-wise and for each block the prediction method is figured based on the proposed method to create prediction errors. According to our prediction method, in each block, 75% of the prediction error pixels can find the ability to embed information. The experimental results show a good acceptable embedding capacity as it is clear in a sample test image of an airplane. In this case, the embedding capacity of 206,270 bits and a Peak Signal-to-Noise Ratio (PSNR) of 50/99 have been reached. These results show the efficiency of our proposed method based on the embedding capacity and visual quality of the images. The outcomes of the proposed method reach better results than the main competitor methods.