Defect depth automation measurement based on image processing for tofd parallel scanning

To get more precise defect depth in parallel scanning of ultrasonic time-of-flight diffraction (TOFD) technique, a method for automatically measuring defect depth using defect image data is proposed in this paper. The method extracts defect image data from B-scan image through a series of image processing to fit a parabolic curve. Defect depth will be easily obtained by combining the quadratic coefficient of fitted parabolic curve with that of formula based on the principle of ultrasonic TOFD parallel scanning. Experiment result shows that the defect depth measuring error is less than 0.5 mm with this method and demonstrates strong practicability, high degree of automation and high measurement precision of the proposed method, reducing ultrasonic testing workload of the operator.