On an image reconstruction method for ECT

摘要

An image by Eddy Current Testing(ECT) is a blurred image to original flaw shape.In order to reconstruct fine flaw image, a new image reconstruction method has been proposed. This method is based on an assumption that a very simple relationship between measured data and source were described by a convolution of response function and flaw shape. This assumption leads to a simple inverse analysis method with deconvolution.In this method, Point Spread Function (PSF) and Line Spread Function(LSF) play a key role in deconvolution processing. This study proposes a simple data processing to determine PSF and LSF from ECT data of machined hole and line flaw. In order to verify its validity, ECT data for SUS316 plate(200×200×l0mm) with artificial machined hole and notch flaw had been acquired by differential coil type sensors(produced by ZETEC Inc).Those data were analyzed by the proposed method. The proposed method restored sharp discrete multiple hole image from interfered data by multiple holes. Also the estimated width of line flaw has been much improved compared with original experimental data. Although proposed inverse analysis strategy is simple and easy to implement,its validity to holes and line flaw have been shown by many results that much finer image than original image have been reconstructed.