Watermark integration to wavelet image coding scheme

With the blossoming demand for efficiently storing and transmitting digital information, image compression has been increasingly vital. On the other hand, at very low bitrates, lossy compression is supposed to be a strong attack to watermarking process. In this work, we attempt to integrate watermarking in wavelet-based image coding. In our study, we used three different wavelet-based compression techniques. The last two coders are respectively included in part I and part II of JPEG2000 standard. Taking advantage of the coding architecture, a watermark embedding is directly performed on the wavelet coefficients before the quantization phase. On the recovery stage, the hide information may be retrieved during the decoding or where ever we decompose the reconstructed image on its wavelet domain. Experimental results have shown that blind watermarking coupled with JPEG2000-based image coders exhibits a high robustness to compression attacks with respect to the quality of the reconstructed images. Moreover its time consuming is negligible compared to coding and decoding duration.