A Location-Map Free Reversible Watermarking With Capacity Control

Reversible watermarking techniques enable the extraction of the embedding bits from a watermarked image in a lossless way. It exploits the high spatial correlation among neighboring pixels. Application in reversible watermarking includes military and medical images. However, images occurs overflow or underflow problems during the imbedding process. Since pixels value may be out of range between [0:255]. Most methods require a location map to solve such problems. In this study, we propose a location map free reversible watermarking algorithm. First, a prediction threshold value is computed, histogram shifting scheme based on the prediction threshold value to solve overflow and underflow problems, second, another threshold value is adopted to achieve capacity control, image quality is better in different payload length with this control. The experimental results reveal that the performance of our proposed method outperforms that proposed by FUJIYOSHI et al. For example, with the same imbedding capacity, the PSNR of our scheme is higher than FUJIYOSHI et al. by 3 dB. Furthermore, our algorithm provides higher embedding capacity compared with FUJIYOSHI et al. Keywords—reversible watermarking; capacity control; predicted value; histogram shifting;