Reversible run length data embedding for medical images
Zhiguo Chang, Jian Xu · 2011
A novel reversible data embedding scheme is proposed, which makes use of the repetitive pixels in a digital medical image to conceal information. The scheme doesn't need the binary location map that is often employed in many reversible data embedding schemes. Performing in the spatial domain without location map compression stage makes the proposed scheme very efficient. Most of information is embedded into the Region of Non-Interest (RONI) and few changes are made to the Region of Interest (ROI), so some preliminary diagnosis can be made without restoration of the original medical image. Experimental results show our scheme can achieve a high data-hiding capacity in many medical images.