Iterative Algorithms for Joint Scatter and Attenuation Estimation From Broken Ray Transform Data

The single-scatter approximation is fundamental in many tomographic imaging problems including x-ray scatter and optical for certain media. In all cases, noisy measurements are affected by both local events nonlocal attenuation. Prior works focus on reconstructing one of two images: density or total However, images media specific useful object identification. Nonlocal effects the attenuation image data summarized broken ray transform (BRT). While analytic inversion formulas exist, poor conditioning inverse problem only exacerbated sampling errors. This has motivated interest related star transforms incorporating BRT from multiple source-detector pairs. methods operate log data. For comprising regions with no a new approach required. We first to present joint estimation algorithm based Poisson models measurement geometry. Monotonic reduction log-likelihood function guaranteed our iterative while alternating updates. also fast computing discrete forward operator. Our generalized can incorporate transmission Transmission resolve low-frequency ambiguity problem, image. benefits estimation, over single-image vary scaling. results quantify these should inform design future acquisition systems.