Register or Log In

International Society for Industrial Process Tomography

8th World Congress on Industrial Process Tomography

Comparison of Industrial Tomography Algorithms for Gamma Scanning 2-D Reconstruction

Hae Yong Kim1, M. I. Haraguchi2, W.A.P. Calvo3

1 - Universidade de São Paulo, Escola Politécnica Av. Prof. Luciano Gualberto, tr. 3, 158 - São Paulo - Brazil hae@lps.usp.br

2 - Tricom Tecnologia Av. Conselheiro Rodrigues Alves, 58 - Piquete - Brazil marcio@tricomtecnologia.com.br

3 - Nuclear and Energy Research Institute – IPEN/CNEN-SP Av. Prof. Lineu Prestes, 2242 - São Paulo - Brazil wapcalvo@ipen.br

ABSTRACT

In this paper, we compare different industrial tomography algorithms to present the result of gamma scanning as a two-dimensional image of density distribution. In two-dimensional gamma scanning, an unconventional irradiation geometry is used and consequently many classic image reconstruction algorithms cannot be used. We tested two iterative reconstruction methods: ART (Algebraic Reconstruction Technique) and MART (Multiplicative Algebraic Reconstruction Technique). The iteration steps of both methods were intercalated with various filters: mean filter, median filter, anisotropic diffusion and total variation. The use of these filters greatly improves the quality of reconstruction. We report the obtained results.

Keywords Tomography Algorithm, Algebraic Reconstruction Technique, Multiplicative Algebraic Reconstruction Technique, Gamma Scanning, Industrial Tomography.

Sign-in to access the full text