Receiver Operating Characteristic (ROC) in Nondestructive Inspection

  • C. Nockemann
  • G.-R. Tillack
  • H. Wessel
  • H. Heidt
  • V. Konchina
Part of the NATO ASI Series book series (NSSE, volume 262)

Abstract

A suitable and objective method to investigate the performance of inspection systems is the Receiver Operating Characteristic (ROC) which is based on the general theory of signal detection. All diagnostic systems include the task of separating a signal from a background of noise. The idea of the ROC method is to characterise the accuracy of an inspection system by evaluating the correct detection rate versus the false alarm rate for a set of different thresholds between signal and noise applied during this task. This method is called detection ROC because it deals only with the pure detection of a signal (defect). If there are several types of defects one can also measure the accuracy of the defect classification and the correctness of the indicated defect importance v/ith the help of the so-called Joint ROC. For the industries it is particularly important to find the unacceptable defects which has to be repaired. For this case the masked ROC is calculated where only the correct or false indications of critical defects are taken into account.

The power of the method is demonstrated in terms of three examples: ultrasonic testing of fibre reinforced plastic materials used in space- and aircraft manufacturing and of welds typical for railway systems and radiographic weld inspection. Of interest was here the dependence of the testing performance from the ultrasonic frequency, from the experience of the inspectors and from the digitisation procedure, respectively. Special tests — reducing the number of samples — were undertaken to show until which minimum amount of statistical basis the result is stable.

Keywords

Receiver Operating Characteristic Receiver Operating Characteristic Curve Inspection System Ultrasonic Testing CFRP Plate 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Metz, C.E. ‘Basic Principles of ROC analysis’. Seminars in Nuclear Medicine 8 4 (1978)CrossRefGoogle Scholar
  2. 2.
    Metz, C.E. ‘Some practical issues of experimental design and data analysis in radiological ROC studies’. Invest Radiol 24 (1989) pp 234-245CrossRefGoogle Scholar
  3. 3.
    Swets, J.A. ‘Assessment of NDT systems-Part I’. Mater Eval 41 11 (1983) pp 1294–1298Google Scholar
  4. 4.
    Swets, J.A. ‘Assessment of NDT systems-Part II’. Mater Eval 41 11 (1983) pp 1299–1303Google Scholar
  5. 5.
    Somoza, E. Mossman, D., and McFeeters, L. ‘The info-ROC technique: A method for comparing and optimizing inspection systems’. Review of Progress in Quantitative Nondestructive Evaluation, vol. 9, ed. D.O. Thompson and D.E. Chimenti, Plenum, New York (1990) pp 601–608Google Scholar
  6. 6.
    Nockemann, C., Heidt, H., and Thomsen, N. ‘Reliability in NDT: ROC study of radiographic weld inspections’. NDT&E International 24 5 (1991) pp 235–245CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media Dordrecht 1994

Authors and Affiliations

  • C. Nockemann
    • 1
  • G.-R. Tillack
    • 1
  • H. Wessel
    • 1
  • H. Heidt
    • 2
  • V. Konchina
    • 3
  1. 1.Bundesanstalt für Materialforschung und -prüfung Laboratory 6.24, Combination and Evaluation of NDT-MethodsBAMBerlinGermany
  2. 2.Materialforschungs und -prüfanstalt an der Hochschule für Architektur und Bauwesen (HAB)MFPAWeimarGermany
  3. 3.St. Petersburg Institute of Railway Transport EngineersPIITSt. PetersburgRussia

Personalised recommendations