Denoising Gray Scale Images Using Curvelet Transform

Digital images are generally of low contrast and they often have a complex type of noise due to various imperfections in acquisition, transmission storage and display devices and also because of application of different types of quantization, reconstruction and enhancement algorithms. The presence of noise gives an image a mottled, grainy, textured or snowy appearance. Image noise comes from a variety of sources. The goal of image denoising is to estimate the original image from noisy image. A number of methods have been developed for denoising of images but most of the methods causes blurring of images and introduces artifacts. In this paper a multiscale curvelet based denoising technique is presented. Image denoising using curvelet transform produce better results than other standard methods used for denoising images. Experimental results show that the Curvelet based method outperforms the median filter and order filter for reducing the additive noise in grayscale images.