期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2016
卷号：5
期号：10
页码：18444
DOI：10.15680/IJIRSET.2016.0510049
出版社：S&S Publications
摘要：Image denoising is one of the basic problems in low level vision.Reduction of noise and enhancing theimages were in spatial domain increases the scope of information in the image. Then, noise and aliasing artifacts areremoved from the structured matrix by applying sparse and low rank matrix decomposition technique. These also helpsin reducing the non-linear artifacts. That is sparsity of the image. Nevertheless, noise amplification and aliasingartifacts are serious in pMRI reconstructed images at high accelerations. Here a low rank matrix decomposition helpsin denoising the medical images using ADMM Algorithm, but was not very much efficient in reducing the error rate.So, redundant multi-resolution decomposition helps in increasing the information levels of the image. And those valueswere shown using performance parameters peak signal noise rate (PSNR) and structural similarity index matrix (SSIM)and entropy i.e, information of an image. And for better accuracy in the pMRI reconstructed images we introduce themethod of continuous convolution process i.e, using intensity gradient vectors. The problem of getting an appropriateabsolute gradient magnitude for edges lies in the method used. The Sobel operator performs a 2-D spatial gradientmeasurement on images. Transferring a 2-D pixel array into statistically uncorrelated data set enhances the removal ofredundant data, as a result, reduction of the amount of data is required to represent a digital image.
