| ABSTRACT: | NEW TECHNIQUES FOR RESOLUTION ENHANCEMENT IN 3D X-RAY TOMOGRAPHIC
IMAGING FROM INCOMPLETE DATA
V.Vengrinovich1, S.Zolotarev1 Y.Denkevich1, and G.-R.Tillack2.
1Institute of Applied Physics, Minsk, Belarus; 2Federal Institute for Materials Research and Testing (BAM);
Berlin; Germany
The objective of this study is to develop algorithms targeting to enhance spatial resolution of 3D x-ray image
reconstruction from limited observations and sparse projections. Limited data tomography is usually
accompanied by restored image blurring and strong artifacts. To overcome the degradation of image
interfaces a hull-like approach is proposed for the description of a surface of an object as well as interior
boundaries. Bayesian image reconstruction technique considering a priori knowledge is adapted for image
description by means of shells. An image correction technique at every iteration step using the hull approach
is developed and tested using both simulated and experimental data. As any integrated noise in this case is
interpreted as a signal, the attempt to extract scattered radiation is applied to decrease the image blurring.
Fusion of data from different reconstruction modalities is worked out to extract the disadvantages of each
modality.
An item analysis is conducted on sets of simulated and real objects having both, binary and multi phase
structure. One of the principle result of this item is that the new algorithms and computer technique for shell
description of an object, integrated into the Bayesian technique for limited data image reconstruction, is
developed and tested. including
The conclusion can be also drawn that the developed technique is capable to be used for corroded pipes wall
thickness assessment having the data from only few x-ray projections.
|