3D SUSAN Filtering of CT Images With Adaptive Selection of the Brightness Threshold

Abstract

In this paper a new application of the Smallest Univalue Segment Assimilating Nucleus (SUSAN) filter is presented aimed at reducing the Additive White Gaussian Noise (AWGN) in Computed Tomography (CT) images. The original 2D implementation of the filter is extended in 3D and scheme for adaptive selection of the brightness threshold is proposed. Testing over large set of CT images reveal its applicability for the purpose and better performance than the 3D Gaussian and Average filters. The current modification of the 3D SUSAN filter is considered also applicable for filtering of other types of 3D images, such as MRI, multispectral and others.