Iterative method for parametric image estimation in MIR

摘要: A new X-ray imaging method called multiple-image radiography (MIR), can simultaneously produce parametric images of absorption, refraction, and ultra-small-angle scatter, while rejecting higher-angle scatter. These where shown to have a superb diagnostic abilities compared some conventional methods (e.g. mammograms). In this paper we present image estimation based on physical model formation in MIR. Presented uses the fact that angular profile measured X-ray, after interaction with object, be modeled as convolution Gaussian curve intrinsic system profile. Further, approach an iterative conjugate gradients algorithm which combination model, offers improvement accuracy.