Two-dimensional temperature distribution measurement by an infrared 2-band emission CT employing CO{sub}2(algorithm of CT when object gas is optically thick)

摘要

The authors propose the CT algorithm for infrared two-band emission CT when optical depth is not so great and Lambert-Beer's Law is applicable. Ordinary CT algorithm like the convolution backprojection method can be used to solve the linear form of the governing equations which appear in the integral of physical values to be obtained. This algorithm can solve the problem of self absorption which was a big obstacle for emission-type CT algorithms. However, when the optical depth is great, Lambert-Beer's Law can not be applied. In this paper, the CT algorithm which can solve the problem of self absorption employing a statistical model for thick optical length is presented. The algorithm is one of the conversion methods and it is verified by simulated projection data for uniform and non-uniform temperature fields. The reconstructed temperature distributions showed very good correspondence with the original distributions. The effect of the noise of detectors on the accuracy of reconstruction is also discussed.