You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither SPIE nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the SPIE website.
12 March 2009Optimal binary coding matrices for multiplexed x-ray imaging
We study a class of binary matrices with excellent noise properties for controlling multiplexing patterns in a multiple-source,
single-detector X-ray system. For such a system, turning multiple sources on at a time in a prescribed pattern
(multiplexing), can offer noise advantages under certain conditions. The patterns used can be represented by binary
matrices which determine the noise properties of the decoded images. Hadamard S-matrices have long been used in
spectroscopy, but they are optimal only in systems with little photonic noise. In X-ray systems with energy-integrating
detectors, the noise structure may be a mix of constant (electronic) noise and noise proportional to the signal (photonic
noise). Under mixed noise conditions, we demonstrate that a certain class of balanced incomplete block design (BIBD)
matrices offers better noise performance over a wider range of noise mixes than the Hadamard matrices. Symmetric
BIBD matrices are characterized by three parameters: v = the number of sources in the multiplexing array; k = the
number of sources on at a time; and λ = the number of multiplexed frames shared by any pair of sources. We compare
noise performance in decoded images for several families of BIBD matrices and show that the BIBD matrices with λ = 1
offer the best performance. We also offer insight into how the available matrices affect parameters of system design in a
multiplexing X-ray system. We conclude that the BIBD(v,k,λ = 1) matrices or matrices derived from them are the best
choices for multiplexing in multi-source X-ray systems.
David S. Lalush
"Optimal binary coding matrices for multiplexed x-ray imaging", Proc. SPIE 7258, Medical Imaging 2009: Physics of Medical Imaging, 72581S (12 March 2009); https://doi.org/10.1117/12.811124
The alert did not successfully save. Please try again later.
David S. Lalush, "Optimal binary coding matrices for multiplexed x-ray imaging," Proc. SPIE 7258, Medical Imaging 2009: Physics of Medical Imaging, 72581S (12 March 2009); https://doi.org/10.1117/12.811124