In the study of optical testing, the computed tomogaphy technique has been widely adopted to reconstruct three-dimensional distributions of physical parameters of various kinds of fluid fields, such as flame, plasma, etc. In most cases, projection data are often stained by noise due to environmental disturbance, instrumental inaccuracy, and other random interruptions. To improve the reconstruction performance in noisy cases, an algorithm that combines a self-adaptive prefiltering denoising approach (SPDA) with a multicriterion iterative reconstruction (MCIR) is proposed and studied. First, the level of noise is approximately estimated with a frequency domain statistical method. Then the cutoff frequency of a Butterworth low-pass filter was established based on the evaluated noise energy. After the SPDA processing, the MCIR algorithm was adopted for limited-view optical computed tomography reconstruction. Simulated reconstruction of two test phantoms and a flame emission spectral tomography experiment were employed to evaluate the performance of SPDA-MCIR in noisy cases. Comparison with some traditional methods and experiment results showed that the SPDA-MCIR combination had obvious improvement in the case of noisy data reconstructions.
© 2007 Optical Society of America
Original Manuscript: August 11, 2006
Revised Manuscript: October 5, 2006
Manuscript Accepted: October 17, 2006
Published: February 20, 2007
Vol. 2, Iss. 4 Virtual Journal for Biomedical Optics
Xiong Wan and Aihan Yin, "Denoising multicriterion iterative reconstruction in emission spectral tomography," Appl. Opt. 46, 1223-1232 (2007)