A Joint Method Based on Wavelet and Curvelet Transform for Image Denoising
Wavelet transform is widely used and has good effect on image denoising.Wavelet transform has unique advantages in dealing with the smooth area of image but is not so perfect in high frequency areas such as the edges.However,curvelet transform can supply this gap when dealing with the high frequency areas because of the characteristic of anisotropy.In this paper,we proposed a new method which is based on the combination of wavelet transform and curvelet transform.Firstly,we detected the edges of the noisy-image using wavelet transform.Based on the edges we divided the image into two parts:the smoothness and the edges.Then,we used different transform methods to dispose different areas of the image,wavelet threshold denoising is used in smoothness while FDCT denoising is used in edges.Experimental results showed that we could get better visual effect and higher PSNR,which indicated that the proposed method is better than using wavelet transform or curvelet transform respectively.
image denoising wavelet transform fast discrete curvelet transform(FDCT) image division PNSR
Hua Wang Jian-zhong Cao Li-nao Tang Zuo-feng Zhou
Xian Institute of Optics and Precision MechanicsChinese Academy of Sciences Xian, 710119, P.R.China
国际会议
西安
英文
758-762
2012-08-24（万方平台首次上网日期，不代表论文的发表时间）