Neural network correction of ultrasonic C-scan images

A neural network-based approach to the correction of C-scan images is presented. This allows the effects of a finite ultrasonic beam diameter in an immersion experiment to be considered, by training the network on a series of defects of known characteristics. The result is an image which is a better representation of the actual defect. (C) 1999 Elsevier Science B.V. All rights reserved