Information

Abstract

Three-dimensional X-ray computed tomography (3DXCT) is a powerful technique for generating a digital 3D volumetric representation of a specimen from a series of 2D X-ray penetration images. The main advantage of 3DXCT is its ability to detect both the interior and the exterior structure of a specimen in one single scan. Having been used in medical diagnostics for a long time, 3DXCT is increasingly employed in industry as a method for nondestructive testing and quality control. One especially challenging industrial application is metrology, which has to fulfill the demands of today’s standards in industrial quality control. 3DXCT facilitates dimensional measurements of internal structures and of inaccessible parts of a component. However the successful industrial application of 3DXCT is constrained by a set of major problems: Artifacts: Industrial 3DXCT systems face problems due to various types of artifacts. The appearance of artifacts in the 3DXCT scan data distorts its correlation to the actual evaluated industrial object and can lead to errors in measurements and false analysis results. Some types of artifacts are affected by the placement of a specimen in the scanning device. Multi-material components: Another problem is occurring when multi-material components (MMCs) are inspected using industrial 3DXCT. Common industrial MMCs may contain metal parts surrounded by plastic materials. A major problem of this type of components is the presence of metal-caused streaking artifacts and distortions. They are located around metal components and significantly influence the material characterization. Furthermore these streaking artefacts and distortions may even prevent any further analysis (especially for the plastic components). Measurements uncertainty: If metrology using 3DXCT is performed, the location of the specimen surface is estimated using the reconstructed 3D volume data. As opposed to mechanical or optical measurement techniques, the surface is not explicit and has a particular positional uncertainty depending on the artifacts and noise in the scan data and the surface extraction algorithm. Conventional CT metrology software does not account for the uncertainty of the data. This thesis is devoted to the development of techniques overcoming the aforementioned problems of common industrial tasks involving the usage of 3DXCT for nondestructive testing and quality control with a main focus on industrial 3DXCT metrology. Several novel contributions utilizing visualization techniques and visual analysis methods were implemented in integrated tools assisting typical industrial 3DXCT tasks during different stages of the data pipeline.

Additional Files and Images

Additional images and videos

teaser: teaser teaser: teaser

Additional files

Weblinks

No further information available.

BibTeX

@phdthesis{amirkhanov-2012-thesis,
  title =      "Visualization of Industrial 3DXCT Data",
  author =     "Artem Amirkhanov",
  year =       "2012",
  abstract =   "Three-dimensional X-ray computed tomography (3DXCT) is a
               powerful technique for generating a digital 3D volumetric
               representation of a specimen from a series of 2D X-ray
               penetration images. The main advantage of 3DXCT is its
               ability to detect both the interior and the exterior
               structure of a specimen in one single scan. Having been used
               in medical diagnostics for a long time, 3DXCT is
               increasingly employed in industry as a method for
               nondestructive testing and quality control. One especially
               challenging industrial application is metrology, which has
               to fulfill the demands of today’s standards in industrial
               quality control. 3DXCT facilitates dimensional measurements
               of internal structures and of inaccessible parts of a
               component. However the successful industrial application of
               3DXCT is constrained by a set of major problems:  Artifacts:
               Industrial 3DXCT systems face problems due to various types
               of artifacts. The appearance of artifacts in the 3DXCT scan
               data distorts its correlation to the actual evaluated
               industrial object and can lead to errors in measurements and
               false analysis results. Some types of artifacts are affected
               by the placement of a specimen in the scanning device. 
               Multi-material components: Another problem is occurring when
               multi-material components (MMCs) are inspected using
               industrial 3DXCT. Common industrial MMCs may contain metal
               parts surrounded by plastic materials. A major problem of
               this type of components is the presence of metal-caused
               streaking artifacts and distortions. They are located around
               metal components and significantly influence the material
               characterization. Furthermore these streaking artefacts and
               distortions may even prevent any further analysis
               (especially for the plastic components). Measurements
               uncertainty: If metrology using 3DXCT is performed, the
               location of the specimen surface is estimated using the
               reconstructed 3D volume data. As opposed to mechanical or
               optical measurement techniques, the surface is not explicit
               and has a particular positional uncertainty depending on the
               artifacts and noise in the scan data and the surface
               extraction algorithm. Conventional CT metrology software
               does not account for the uncertainty of the data. This
               thesis is devoted to the development of techniques
               overcoming the aforementioned problems of common industrial
               tasks involving the usage of 3DXCT for nondestructive
               testing and quality control with a main focus on industrial
               3DXCT metrology. Several novel contributions utilizing
               visualization techniques and visual analysis methods were
               implemented in integrated tools assisting typical industrial
               3DXCT tasks during different stages of the data pipeline.",
  month =      nov,
  address =    "Favoritenstrasse 9-11/E193-02, A-1040 Vienna, Austria",
  school =     "Institute of Computer Graphics and Algorithms, Vienna
               University of Technology ",
  URL =        "https://www.cg.tuwien.ac.at/research/publications/2012/amirkhanov-2012-thesis/",
}