A Novel Framework for Robust Lossless Data Hiding

Histogram-based lossless data hiding (LDH) has been recognized as an effective and efficient way in the field of data hiding. Recently, a LDH method using the statistical quantity histogram (SQH) has been a hotspot for its good performance, which can extract secret correctly even if stego-image has some extent distortion. However, this method has some shortcoming which limits its application in practice. For this purpose, we develop a novel framework, which consists of the following three components:1) In order to overcome the shortcomings of traditional SQH, proposing a pre-processing method to construct a suitable carrier for information embedding; 2)Applying LDH method to restore the bitmap information ,which will be used for reversibly recovering the host image; 3) Presenting a robust lossless hiding algorithm, which guarantees host images can be recovered losslessly in the case of stego-image remains intact, on the contrary, the secret information can also be resistant to a certain degree of JPEG2000 attacks. Thorough experiments over different kinds of images demonstrate the effectiveness of the proposed framework.