Terahertz computed tomographic reconstruction and its wavelet-based segmentation by fusion

Abstract

In this paper, terahertz (T-ray) computed tomographic (CT) imaging [1] and segmentation techniques are investigated. The traditional filtered back projection is applied for the reconstruction of terahertz coherent tomography. A set of linear image fusion and novel wavelet scale correlation segmentation techniques is adopted to achieve material discrimination within a three dimensional (3D) object. The methods are applied to a T-ray CT image dataset taken from a plastic vial containing a plastic tube. This setup simulates the imaging of a simple nested organic structure, which provides an indication of the potential for using T-ray CT imaging to achieve T-ray pulsed signal classification of heterogeneous layers. The wavelet based fusion scheme enjoys the additional benefit that it does not require the calculation of a single threshold and there is only a segmenter parameter to adjust. ©2007 IEEE.