Noise removal for medical X-ray images in wavelet domain

Abstract

Many important problems in engineering and science are well-modeled by Poisson noise, and the noise of medical X-ray images is Poisson noise. In this paper, we propose a method for noise removal for degraded medical X-ray images using improved preprocessing and an improved BayesShrink (IBS) method in the wavelet domain. First, we preprocess the medical X-ray image. Second, we apply the Daubechies (db) wavelet transform to medical X-ray images to acquire scaling and wavelet coefficients. Third, we apply the proposed IBS method to process wavelet coefficients. Finally, we compute the inverse wavelet transform for the threshold coefficients. Experimental results show that the proposed method always outperforms traditional methods. © 2008 Wiley Periodicals, Inc.