An adaptive reversible data hiding scheme based on histogram shifting using decimal floating signed-digit stream

This paper proposes a novel method based on histogram shifting using signed digits for data hiding. Our proposed method takes the prediction errors obtained from the original image using a 4×4 block-wise prediction. Then, we embed the information in the prediction errors of the image using the histogram shifting technique. A crucial point regarding the embeddable data in this method is that we divide the binary stream into equal parts of two, three, or four bits. For each two, three, or four-bit digit, we consider a numerical equivalent using the approach described in this paper. Subsequently, based on each of the signed digits, assigned floating numbers are used to represent the embeddable information instead of the binary stream. Experimental results for a sample image, "Airplane", with four-bit data segmentation demonstrate an outstanding embedding capacity of 825,080 bits and a PSNR of 33.87 dB, indicating that our proposed scheme achieves a remarkably high embedding capacity while maintaining an acceptable visual quality.

An adaptive reversible data hiding scheme based on histogram shifting using decimal floating signed-digit stream Keywords:

Reversible data hiding , Histogram shifting , Prediction error , Signed digit stream , Floating numbers , Embedding capacity , Visual quality