Title :
Artificial intelligence techniques for the automated analysis of ultrasonic NDT data
Author :
McNab, A. ; Dunlop, Iain
Abstract :
The following methods of NDT data processing techniques are considered: removal of noisy or spurious indications; derivation of missing indications; smoothing; thresholding; edge detection and flood-filling; and contouring. Classification of suspect regions and shape recognition are discussed. The following types of automated recognition techniques are considered: best-fit analysis; and triangulation analysis
Conference_Titel :
Advances in Transducers, Equipment and Data Processing to Improve the Reliability of NDT, IEE Colloquium on
Conference_Location :
London