Description

ABSTRACT   

                     Here we propose the novel method to reduce the noise from the image.  Initially we

have to generate the Gaussian noise to the input image. To remove the noise we use the Shearlet

transform. Shearlets are a multiscale framework which allows to efficiently encoding anisotropic

features in multivariate problem classes. The shearlet transform is unlike the traditional wavelet

transform which does not posses the ability to detect directionality. After that we apply the

shrinkage function. Shrinkage is also called thresholding. Here we predict the noise threshold in

the shearlet co-efficient. After that it will be removed from the co-efficient. After shrinkage

function it is passed to the inverse transform to reconstruct the image. In that image we calculate

the MSE, PSNR value for the analysis. After shearlet denoising, it passed to G-map filter to

increase the image quality.