Reversible data hiding using reference points coding

This paper proposed the reference points coding method for reversible data embedding. The proposed method provides the user with the flexibility to choose the level of secret data embedding from light embedded to heavier embedding depending on either the size of a certain secret image or that of level of secrecy required during transfer. This is achieved by first comparing each pixel value to that of its neighboring pixels, when same values are found, its own pixel value will be replaced by a two bit reference address of that neighboring pixel. The remaining space of that pixel is then ready for data embedding. The user then may choose to embed up to three bits per available pixel while maintaining good levels of PSNR. For added security, the user may choose to embed a varied numbers of bits per available pixel. This practice would lead to better protection against detection and steganalysis compared to fixed number embedding methods.