[14a-2E-1~8] 3.3 Information photonics and image engineering

[14a-2E-7] Improvement in accuracy of the reconstruction of computed tomography using moving average for an incomplete projection data set with noise

〇Kodai Kurita¹, Satoshi Tomioka¹, Shusuke Nishiyama¹ (1.Hokkaido Univ.)

Keywords:interferometry,tomography,measurement of refractive index distribution

By combining an interferometer and a computed tomography, a 3-dimensional non-destructive temperature measurement around flame may be realized. In this observation system, however, the incident beam angle is limited due to the mirror arrangement. As the result, the projection data set is incomplete. Moreover, the signal to noise ratio is low, since the projection data are obtain with short exposure time to avoid effects of vibration of mirrors caused by the movements of mirrors. In order to overcome these problems, we propose a reconstruction algorithm having the robustness against the noise by adding a moving average scheme to the maximum likelihood expectation maximization (ML-EM) algorithm.