Denoising of Gaussian and Speckle Noise from XRay Scans using Haar Wavelet Transform

Abstract

This paper presents an algorithm for reducing speckle and Gaussian noise from medical X-Ray scans. Haar wavelet analysis has been applied to eliminate noise while preserving the sharpness of salient features. Both the noise forms are augmented in the input xray scans. The level of 30% noise is added into the input X-ray image. The image is decomposed upto level 2 using soft thresholding technique. The approach for speckle noise reduction is shown to be more effective than that affected by Gaussian noise. A study using a clinical X-Ray image suggests that such denoising and enhancement may improve the overall consistency of expert observers to manually defined borders