During computed tomography (CT), a set of images is taken from different angles, covering usually one full rotation (360°). These images will be called projections in the further text. The set typically consists of hundreds to thousands of projections. Of course, exposition of such an amount of projections takes a significant time (typically in order of tens of minutes to units of hours) and during this time, there can be drift in electrical (centering currents) as well as mechanical parameters (temperature effects, wearing out of the target) of the X-ray tube, manifesting themselves as a geometrical movement of the X-ray source with respect to the rotational axis. This movement can be significant in particular with high magnification, i.e., high-resolution micro CT, when there is a big ratio of the distances tube - detector and tube - object. The effect of tube spot movement is then observable in the projections basically as a shift of the rotational axis, although the real distortion caused by the moved spot, i.e., changed setup geometry, is more complicated. The observation of the shift is possible in two projections made under the same angle.
The tube spot movement during acquisition of the projections affects the sharpness of the 3D reconstruction considerably, which of course leads to lower resolution then theoretically possible with the given magnification.
It is therefore of a great interest to compensate the tube spot movement.
ToraSpot is a software that facilitates calculation of shifts of reference images and perform the shifting of the projections in individual sets. It has been developed in the Centre of Excellence Telč and can be used for any sets of data acquiered in the way described above. The software has its own graphical user interface (GUI) made in PyQt5. The core algorithms are implemented in Python 2.7.
Getting the software and Licensing
Software files can be obtained from assoc. prof. Michal Vopálenský, firstname.lastname@example.org. The software is distributed for free, but we require referencing when used for the data treatment in a publication. Please, involve the sentene "To eliminate the influence of the spot movement, data were processed with the ToraSpot software, developed in the Centre of Excellence Telč, Institute of Theoretical and Applied Mechanics, Czech Academy of Sciences."
System requirements and installation
ToraSpot was developed for computers using Windows operating system. Its usage on computers with different operating systems should be possible, but was not tested. The software requires Python 2.7 with PyQt5, numpy and scipy.ndimage packages installed on the computer. The whole software package consists of four files: progressWindow.ui, SMCForm.ui, SMCWidget.ui and SMC.py. The .ui files are the components of the GUI, SMC.py is the core program containing 668 lines of Python code. All these files have to be located in one directory.
The software does not require installation, it can be launched by executing SMC.py. Dependent on the platform, this may require a simple doubleclick, or it has to be launched from the commander line going to the program directory and typing python SMC.py If everything goes well, the console commander line will start up together with the basic GUI face of ToraSpot.
The result was achieved with the financial support of the Ministry of Education, Youth and Sports of the Czech Republic under the NPU I programme, project LO1219.