Lossless embedding using pixel differences and histogram shifting technique

Abstract

Generally hiding information destroys the host image, however, some sensitive images for which any embedding distortion of the image is intolerable, such as military images, medical images, or artwork preservation etc. Reversible data hiding techniques are designed to solve the problem of Lossless embedding of large messages in digital images so that after the embedded message is extracted, the image can be completely restored to its original state before embedding occurred. But the solutions came with histogram-based reversible data hiding techniques, in which the message is embedded into the histogram bin. However, in those techniques the recipients are communicated pairs of peak and zero points. By extending to the histogram modification technique using pixel differences hiding capacity can be increased and using a binary tree structure requirement to communicate pairs of peak and zero points to the recipient is eliminated. By adopting a histogram shifting technique overflow and underflow can be prevented. Neighbor pixels are often highly correlated and have spatial redundancy; the differences have a Laplacian-like distribution. This enables large Hiding capacity increased while keeping embedding distortion low. The results show using pixel differences & histogram shifting, increase in PSNR and decrease in MSE is seen as compared to histogram based reversible data hiding technique.