An Infrared Thermal Image Defect Detection Method Based on Phase Features*.

The typical eddy current thermal imaging data processing method-Fourier transform has the problems of long time and low efficiency. To improve these defects, a fast phase calculation method is used for detection. The phase expansion is performed on the temperature sequence composed of multiple continuous heat maps through the fast Fourie transform formula, and the phase calculation algorithm related to the conversion period is implemented. In the experiment, when the sampling point is 700, the efficiency is increased by 2.43 times. Through the analysis of the defective steel plate thermal image sequence, a clear display of defects and defect edges is realized, which verifies that the method is effective and accurate, and can greatly improve the ability of defect edge detection.