An artificial neural network interpreter for transient thermography image data

It is shown that an artificial neural network can be used as an expert system to obtain details of defect size and depth from transient thermography data. The neural network was trained on a comprehensive range of finite difference modelling results and is shown to provide a simple means of interpreting practical experimental measurements. An assessment is given of the sensitivity of the method to likely uncertainties in experimental measurement. (C) 1997 Elsevier Science Ltd.