CT Colonoscopy Medical Images Enhancement By Wavelet Threshold Estimation and Nonlinear Exterapolation Filter with Dosimetry Consideration

CT Colonoscopy Imaging have seen growing interest in recent years. Recently, The CT dosimetry has been become a heated controversial issue in other words the imposed dose to patient due to CT imaging is too important. The use of CT for the procedure impels limiting th dose over time. Image sequences obtained thereby exhibit high level of noise and very low contrasts. Hence , the development of efficient methods to enable optimal visualization of these sequences is crucial. In this paper we propose a denoising method based on the wavelet transform. First , we apply a new soft threshold based on modify bessel function to high frequency coefficients of wavelet and on the other hands apply a local adaptive wiener filter to low frequency coefficients of wavelet. Finally, a third step is carried out to enhance feature of image specially edges of image. For this, we propose a edges reinforcement algorithm by nonlinear exterapolation filter. We used CT Colonoscopy images for denoising .For this purpose, Gaussian noise was added to the images because of the fact that actual noise of the CT colonoscopy images derived from radiation scattering in the detectors of CT scan system resembles Gaussian function. Images quality enhancement was quantified using signal to peak noise ratio (PSNR) and the images edges sharpening was evaluated by the images structural similarity (Ssim) criterion. The benefit and dominance of this technique was confirmed compared to other techniques of Wavelet thresholding such as VisueShrink,Sureshrink, BayesShrink, AdaptShrink, LAWMAP and NeighShrink.

Keywords:

Computed Tomography Colonoscopy (CT) , Image Denoising , structural similarity , signal to peak noise ratio , Wavelet threshold