An evaluation of three algorithms for reconstructing images from data with missing projections

Abstract

Three iterative reconstruction algorithm for reconstructing images from tomographic data with missing projections are evaluated. One algorithm iteratively minimizes a squared-error criteria; the other two, both transform algorithms, estimate the missing projections by placing constraints on the missing data. These algorithms are evaluated on the basis of their accuracy and computational speed. The transform methods are shown to be both faster and more accurate than the least-squares method.