Method for crack tip recognition in an ultrasonic time of flight diffraction image

An aluminum sample containing artificial internal crack is inspected by using the method of ultrasonic time of flight diffraction (TOFD), and the low resolution B-scan image with hyperbolic feature is processed in order to improve testing precision. According to the inspection manner of ultrasonic TOFD, a synthetic aperture focusing technique (SAFT) algorithmic model for B-scan image reconstruction is founded. By using the model, ultrasonic TOFD B-scan image can be enhanced. The time resolution of the B-scan image is improved by the pre-processing of searching for gray peak value, and the lateral resolution of the image is enhanced by suppression of the cutter by the following processing of SAFT. A B-scan image processing method that based on searching for gray peak value and SAFT is therefore proposed. A processing window is adopted in order to minimize the execution time of the algorithm, and proper width of the window is chosen. The result shows that the method can effectively enhance the target in the image, and the location of upper and lower crack tip in the sample can be measured rapidly and accurately accordingly.