Efficient Binary Tomographic Reconstruction

Authors: Stéphane Roux Hugo Leclerc François Hild

Publish Date: 2013/09/18

Volume: 49, Issue: 2, Pages: 335-351

PDF Link

Abstract

Tomographic reconstruction of a binary image from few projections is considered A novel heuristic algorithm is proposed the central element of which is a nonlinear transformation ψp=logp/1−p of the probability p that a pixel of the sought image be 1valued It consists of backprojections based on ψp and iterative corrections Application of this algorithm to a series of artificial test cases leads to exact binary reconstructions ie recovery of the binary image for each single pixel from the knowledge of projection data over a few directions Images up to 106 pixels are reconstructed in a few seconds A series of test cases is performed for comparison with previous methods showing a better efficiency and reduced computation timesCommunication of the raw tomographic data shown in Fig 4 by E Gouillart CNRS/SaintGobain Aubervilliers France and C Zang RWTH Aachen Germany is gratefully acknowledged We are also indebted to E Gouillart for the suggestion that boundary sites may be an appropriate measure of complexity as proposed in Ref 24 We also thank an anonymous reviewer for interesting and constructive remarks This work is supported by the French Agence Nationale de la Recherche through “RUPXCUBE” ANR09BLAN000901 and “EDDAM” ANR11BS09027 projects

Keywords: