Curves-Based Similarity Method (CBSM) for Defect Depth Quantization

• Published in: IEEE transactions on instrumentation and measurement Vol. 72; p. 1
• Main Authors: Yang, Pengbin, Zhou, Xiuyun, He, Ruijie, Liu, Zhen
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.01.2023; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Algorithms; Cooling rate; Defect Detection; Defects; Depth Quantification; Eddy currents; Feature extraction; Image segmentation; Image sequences; Inspection; Principal components analysis; Quantization (signal); Similarity; Surface impedance; Thermal imaging; Thermography
• ISSN: 0018-9456; 1557-9662
• DOI: 10.1109/TIM.2023.3323993

In the pulse eddy current thermal imaging experiments, the trends of temperature response curves of buried defects at different depths are same, but there are differences in cooling rates. The accuracy of depth quantification of buried defects can be improved by making full use of the rich information contained in the temperature response curves. To this end, a feature extraction, defect segmentation and depth quantification algorithm named curves-based similarity method is proposed in this paper. By making comprehensive use of the global and local features of thermal image sequences, the average similarity of the temperature response curve is used as the feature extraction method and quantification parameter. The effectiveness of the method is verified by simulation and experiment. The results show that the method can better enhance the defect information and suppress the noise compared with PCA (Principal Component Analysis) and PPT (Pulsed Phase Thermography). Additionally, the average error of the quantization results of the algorithm is reduced by 1.33% compared with the characteristic time quantization method.