Wavelet Packet Based CT Image Denoising Using Bilateral Method and Bayes Shrinkage Rule

Abstract

The reconstruction process of Computed tomography (CT) is a typical task to get the CT images. There are certain processes to reconstruct the CT images. X-rays are transmitted to human body; raw data is collected by detectors over the different directions, and finally radon and inverse radon transform has been performed to reconstruct the CT images. The efficiency of whole process is important task. But due to software, hardware and other transmissions and mathematical problems the noise may appears in CT images. The reason of noise appear in CT image is also depend on the X-rays transmission. If the higher amount of X-rays are transmitted over the human body organs, then the quality of CT images are good but it may not good for affect human body organs. With low amount of CT images, human body organs may safe but noise is degraded to the quality of CT images. Hence, if noise can be suppressed from low dose CT images, it will good for the society. Various research have been already done to reduce noise from the CT images, still it is a challenging task. To suppress noise from CT images, three major techniques are categorized: Projection based denoising, Iterative based denoising and Post processing based denoising. In projection based denoising, the CT images are filtered when CT images are reconstructed through projected X-ray beams such as filtering of sinogram using bilateral filtering over the low dose CT images [1–3]. In iterative based denoising, CT images are reconstructed using an iterative approach such as iterative CT image reconstruction using shearlet transform [4, 5]