How Data Processing Influences Quantification of Internal Features of Known and Unknown Sizes: Effect of User Experience and Software

  • Chris Packer (Speaker)
  • Tawfik, A. (Contributor to Paper or Presentation)
  • Addinall, K. (Contributor to Paper or Presentation)
  • F. Leonard (Contributor to Paper or Presentation)
  • J. Ahuir Torres (Contributor to Paper or Presentation)
  • M. Sharp (Contributor to Paper or Presentation)
  • Blunt, L. (Contributor to Paper or Presentation)
  • Younes Chahid (Contributor to Paper or Presentation)
  • P. Sarvade (Contributor to Paper or Presentation)
  • I. Lister (Contributor to Paper or Presentation)
  • C. Atkins (Contributor to Paper or Presentation)
  • S. Tammas-Williams (Speaker)

Activity: Talk or presentation typesOral presentation

Description

While XCT can provide unprecedented access to the internal structure of material and components, the processing route used to turn greyscale data into numerical measurements can vary significantly. A review of the literature reveals that while XCT radiography settings are often well reported, details regarding the methods used to process the reconstructed data are not. While not possible to rectify this issue, here we quantify the accuracy and precision of measurements made by XCT data analysists from a range of backgrounds (undergraduate to professional) using different software packages with
varying levels of process automation. To provide meaningful results, all contributors used the same data. XCT was conducted on an object deliberately populated with 300 voids between 10 and 100 µm in diameter, using laser micro-drilling, which can be made fully internal by assembly of the object. Non-XCT measurement techniques were used to measure the voids, to provide a “true” value for contrast to that measured by XCT. Finally, an additively manufactured object containing unknown sized internal features is analysed by the same users, and the results compared to indirect measurement of the porosity via vacuum outgassing to test whether the differences in measurement could lead to components being accepted/rejected incorrectly
Period16 Jun 2022
Event title6th Dimensional X-ray Computed Tomography Conference
Event typeConference
Conference number6
LocationManchester, United Kingdom
Degree of RecognitionInternational