Statistical Estimation Theory Approach for the Dynamic Image Reconstruction

G.-R.Tillack1, V. M. Artemiev2, and A. O. Naumov2

1 Federal Institute for Materials Research and Testing (BAM), Unter den Eichen 87, 12205 Berlin, Germany, Gerd-Ruediger.Tillack@bam.de

2 Institute of Applied Physics, Akademicheskaya str.16, 220072, Minsk, Belarus, artemiev@iaph.bas-net.by, naumov@iaph.bas-net.by

ABSTRACT

The process tomography imaging considers two main tasks. The first one can be interpreted as an image reconstruction when the object is moving relatively to the acquisition setup. The second one is a reconstruction of an object with time varying internal properties. Both of these cases lead to solu- tions of the image reconstruction problem that take in account varying in time image properties. The paper presents the statistical estimation theory approach for the dynamic image reconstruction algo- rithms development. The image model is supposed to be a dynamic random field discrete in time and space. For the task of reconstructing this dynamic field a Kalman filter technique in the state space is proposed. The applicability of this approach to the dynamic reconstruction task is shown by examples for simulated data.

Keywords dynamic image, stochastic estimation, dynamic reconstruction, Kalman filter