AN INTELLIGENT IMAGING SCHEME FOR AUTOMATED EDDY CURRENT TESTING

An intelligent imaging scheme involving artificial neural networks and image processing methods is presented for automated eddy current testing of austenitic stainless steel plates. A quick coarse-scan imaging is performed to identify all the defective regions, which are later subjected to detailed fine-scan imaging to produce accurate threedimensional pictures of the defects. Experimental studies confirm that the imaging scheme is insensitive to disturbing variables and is able to significantly reduce the imaging time as well as the data storage requirements. Details of the scheme and its applications are given.