skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Systematic Approach for Validation of X-Ray Automatic Defect Recognition Systems

Abstract

With the advent of digital radiography, there has been a gradual shift from operators viewing images to find defects to totally automated defect recognition (ADR) systems. This has resulted in reduced operator subjectivity, reduced operator fatigue, and increased productivity. These automated defect recognition solutions are based on reference or non-reference based approaches or a combination of both. There exists some amount of uncertainty or reluctance to accept automated systems in view of no systematic quantified metrics available on performance of these ADR systems in comparison to human operators. This paper describes the metrics that one could follow to quantify the performance of ADR systems such as detectability for different defect types and sizes, accuracy, false call rate, robustness to various noise levels etc., As it might be difficult to have images with defects of various sizes, shapes, contrast and noise levels, a methodology to generate images with simulated defects with variability's in parameters such as size, shape, contrast to noise ratio etc., is demonstrated. This can be used to generate probability of detection estimates for different defect types and geometries. This will result in establishing confidence limits for ADR systems and can be used to judge if it would meetmore » specific customer requirement. This would facilitate increase in the acceptability of ADR systems over current manual defect recognition systems for applications in various industries such as Castings, Oil and Gas, Aviation etc.« less

Authors:
; ; ;  [1]
  1. John F. Welch Technology Centre, GE Global Research Centre, Bangalore (India)
Publication Date:
OSTI Identifier:
21054958
Resource Type:
Journal Article
Resource Relation:
Journal Name: AIP Conference Proceedings; Journal Volume: 894; Journal Issue: 1; Conference: Conference on review of progress in quantitative nondestructive evaluation, Portland, OR (United States), 30 Jul - 4 Aug 2006; Other Information: DOI: 10.1063/1.2718188; (c) 2007 American Institute of Physics; Country of input: International Atomic Energy Agency (IAEA)
Country of Publication:
United States
Language:
English
Subject:
71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSICS; 36 MATERIALS SCIENCE; ACCURACY; COMPARATIVE EVALUATIONS; DEFECTS; DETECTION; FATIGUE; IMAGES; NOISE; PERFORMANCE; PROBABILITY; VALIDATION; X RADIATION; X-RAY RADIOGRAPHY

Citation Formats

Navalgund, Megha, Venkatachalam, Rajashekar, Asati, Mahesh, and Venugopal, Manoharan. Systematic Approach for Validation of X-Ray Automatic Defect Recognition Systems. United States: N. p., 2007. Web. doi:10.1063/1.2718188.
Navalgund, Megha, Venkatachalam, Rajashekar, Asati, Mahesh, & Venugopal, Manoharan. Systematic Approach for Validation of X-Ray Automatic Defect Recognition Systems. United States. doi:10.1063/1.2718188.
Navalgund, Megha, Venkatachalam, Rajashekar, Asati, Mahesh, and Venugopal, Manoharan. Wed . "Systematic Approach for Validation of X-Ray Automatic Defect Recognition Systems". United States. doi:10.1063/1.2718188.
@article{osti_21054958,
title = {Systematic Approach for Validation of X-Ray Automatic Defect Recognition Systems},
author = {Navalgund, Megha and Venkatachalam, Rajashekar and Asati, Mahesh and Venugopal, Manoharan},
abstractNote = {With the advent of digital radiography, there has been a gradual shift from operators viewing images to find defects to totally automated defect recognition (ADR) systems. This has resulted in reduced operator subjectivity, reduced operator fatigue, and increased productivity. These automated defect recognition solutions are based on reference or non-reference based approaches or a combination of both. There exists some amount of uncertainty or reluctance to accept automated systems in view of no systematic quantified metrics available on performance of these ADR systems in comparison to human operators. This paper describes the metrics that one could follow to quantify the performance of ADR systems such as detectability for different defect types and sizes, accuracy, false call rate, robustness to various noise levels etc., As it might be difficult to have images with defects of various sizes, shapes, contrast and noise levels, a methodology to generate images with simulated defects with variability's in parameters such as size, shape, contrast to noise ratio etc., is demonstrated. This can be used to generate probability of detection estimates for different defect types and geometries. This will result in establishing confidence limits for ADR systems and can be used to judge if it would meet specific customer requirement. This would facilitate increase in the acceptability of ADR systems over current manual defect recognition systems for applications in various industries such as Castings, Oil and Gas, Aviation etc.},
doi = {10.1063/1.2718188},
journal = {AIP Conference Proceedings},
number = 1,
volume = 894,
place = {United States},
year = {Wed Mar 21 00:00:00 EDT 2007},
month = {Wed Mar 21 00:00:00 EDT 2007}
}