Reversible Data Hiding using Pixel-Value-Ordering and Difference Expansion in Digital Images

Abstract

Pixel value ordering (PVO) and difference expansion (DE) are promising methods for digital image steganography. PVO is a steganographic method of digital images that deals with pixel values sorting before the data embedding. In digital image steganography, DE is a method for data protection based on hiding secret data in differences computed between pixels. Even though the previous methods based on PVO and DE tried to improve the embedding capacity and the stego image quality, improvement is still needed to increase both simultaneously. This work proposes a new method that combines PVO and DE to improve the number of embeddable areas and the embedding capacity in the pixels of a digital image. The experimental results showed that the maximum number of the embeddable regions within used images with the existing methods was increased from 17582 pixels to 131009 pixels, yielding the embedding capacity improvement from 6.70% up to 49.97%.