The Nonsubsampled Contourlet-Wavelet Hybrid Transform: Design and Application to Image Watermarking

In this paper, we develop nonsubsampled contourlet and wavelet hybrid transform (NSCWHT) and study its applications. The filtering banks construction proposed in this paper is based on the nonsubsampled contourlet transform (NSCT) and wavelet transform (WT). The result is point singularity and line singularity in image can be represented simultaneously. We combine NSCT with WT to design filters that lead to a NSCWHT with better singularity representation when compared to the single one of them. We propose a design framework based on the hybrid approach, that allows for a fast implementation based on NSCT and WT respectively. In addition, we propose a new image watermarking algorithm based on the developed NSCWHT, we assess the performance of the NSCWHT in this image watermarking algorithm. In this application, the NSCWHT compares favorably to other existing methods in the literature.