Removal of Gaussian noise from stationary image using shift invariant wavelet transform
Vikas Gupta, Rajesh Mahadeva, Ashish Shukla · 2013
Discrete wavelet transform (DWT) has gained widespread recognition and popularity in image processing due to its ability of capturing energy of signal in a few energy transform value. As well as it has also ability to underline and represent time-varying spectral properties of many transient and other nonstationary signals. In DWT denoising is done only in detail coefficient, this offer advantage of smoothness and adaption. However DWT has a lack of shift invariance. This shift-variance is a major problem with the use of DWT for transient signal analysis and pattern recognition applications. Denoising of images with the DWT some time also give visual artifacts due to Gibbs phenomena in neighbourhood of discontinuities. In this paper, a shift-invariant analysis scheme is proposed for removing of additive Gaussian noise in stationary image. An investigation has been made on discrete wavelet transform with shift invariant in terms of PSNR and visual performance.