The Use of Multispectral Imaging in Limited-Angle Reconstruction

It is known that some Fourier components of an object are missing from its projection data in CT imaging if the object can be scanned only in a restricted angular range. Due to the ill-conditioning of the problem, the accuracy of extrapolating the missing Fourier components from the known ones is limited by the precision of both the measured data and the computing device used. In this paper we propose a method for improving limited-angle reconstruction by separating out individual component substances in the object through multiple energy scans. It is shown that reconstructing individual component substances can be done more accurately than reconstructing the composite object. The use of nonlinear constraints in reconstruction is discussed. The method can be applied to modalities such as x-ray scanning, electron microscopy, ultrasound imaging, and neutron imaging.