Pseudo Ridgelet Transform for Image Denoising

Wavelet transforms have been successfully used in many scientific fields such as image denoising. Ridgelets is a new system of representations, which deals effectively with line singularities in 2D. However, the discrete version of the ridgelet transform would result in either redundancy or non-reconstruction. In this paper, a pseudo ridgelet transform is developed to overcome this weakness. Since a moving window-based pyramid is employed, the multiscale filter architecture is obtained. Simplicity and effectiveness make it attractive in image denoising.