Reversible Data Hiding with Prediction-based Histogram Alteration

In this paper, we propos e a new algorithm in reversible data hiding, with the applications of copyright and privacy protection for natural image s. Security issues for digital images have long been an important topic in research and practical applications, and reversible data hiding has attracted more and more attention. Unlike conventional schemes, we take the prediction of luminance values and their spatia l coordinates into consideration in order to look for better performances. W e take the inherent characteristics of original image into account, with the predictionbased difference histogram alteration. By doing so, the larg er embedding capacity with similar output image quality in reversible data hiding can be obtained. Simulation results demonstrate the sup eriority over exis ting schemes, and the effectiveness for practical applications for digital images.