Flaw Depth Classification in Eddy Current Tubing Inspection by Using neural Network

摘要

Flaw depth estimation is crucial in eddy current tubing inspection in order to prevent leak accidents in various types of heat exchangers. Udpa proposed a novel method using neural network to classify four different types of flaws detected by eddy current tubing inspection [1, 2]. They used as the neural network input the Fourier descriptor coefficients of cumulative angular function of flaw signal pattern curve [3]. Their classification is based on the shape differences in signal patterns because the coefficients are invariant under rotation, translation, and scaling of the signal pattern.