Suppression of residual gradients in the flat-field corrected images

0507997 - ÚTAM 2020 RIV DE eng C - Conference Paper (international conference)
Vopálenský, Michal - Kumpová, Ivana - Vavřík, Daniel
Suppression of residual gradients in the flat-field corrected images.
9th Conference on Industrial Computed Tomography (iCT) 2019. Bad Breisig: NDT.net, 2019, č. článku 23716. The Web's Largest Open Access Database of Nondestructive Testing (NDT). ISSN 1435-4934.
[Conference on Industrial Computed Tomography (iCT) 2019 /9./. Padova (IT), 13.02.2019-15.02.2019]
EU Projects: European Commission(XE) ATCZ38 - Com3d-XCT
Institutional support: RVO:68378297
Keywords : image processing * flat field correction * radiography * computed tomography * image correction methods
OECD category: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
http://www.ndt.net/?id=23716

Flat field correction, based on using the open beam and the dark field images, is a basic method used for correction of the projection images acquired in computed tomography process. The correction suppresses the stationary noise of the detector, caused by non-uniform gain and offset of each pixel, as well as the non-uniformity of intensity caused by the non-flat intensity profile of the incident X-rays beam. In an ideal case, the corrected image should have a constant intensity in the open-beam areas, i.e., everywhere outside the investigated object. However, this is not always true due to several factors, such as drifts of the spot position, wearing out of the target (pitting), etc. In such a case, the background in the flat-field corrected image is not really “flat” or constant. Particularly in radiography, this is an undesired effect, because the intensity values are also influenced in the object area, leading thus to a possible misinterpretation of the image (even for a constant thickness area, the intensity distribution looks like if the thickness was varying). In this paper, a method is described how to compensate this effect in rather an effective way.
Permanent Link: http://hdl.handle.net/11104/0298968