Title: Dual tree complex wavelet transform incorporating SVD and bilateral filter for image denoising

Authors: Mohan Laavanya; Marappan Karthikeyan

Addresses: Department of ECE, Adithya Institute of Technology, Coimbatore, India ' Tamilnadu College of Engineering, Coimbatore, India

Abstract: In recent years massive production of digital images increased the need for image denoising. The effect of noise can be removed by using spatial and frequency domain approaches. Discrete Wavelet Transforms (DWT) is a frequency domain approach, which removes the noise by shrinking the wavelet coefficients using simple threshold value. Even though wavelet transform is popularly used in image processing applications, shift variance and poor directional selectivity are the two noteworthy limitations. In order to overcome these limitations, Dual Tree Complex Wavelet Transform (DTCWT) is used for perfect reconstruction of noisy image. A DTCWT incorporating Singular Value Decomposition (SVD) with Frobenius energy correcting factor and bilateral filter for image denoising using bivariate shrinkage function for thresholding the image is proposed in this paper. The denoising performance of the proposed method in terms of PSNR and it indicates that the proposed method outperforms over other existing techniques.

Keywords: bilateral filter; bivariate shrinkage; DTCWT; image denoising; SVD; thresholding technique; wavelet transform.

DOI: 10.1504/IJBET.2018.089956

International Journal of Biomedical Engineering and Technology, 2018 Vol.26 No.3/4, pp.266 - 278

Received: 16 Mar 2017
Accepted: 06 Apr 2017
Published online: 26 Feb 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article