Interactive Methodology for Optimized Defect Characterization by Quantitative Pulsed Phase Thermography

ABSTRACT Pulsed phase thermography is a nondestructive evaluation processing technique based on the discrete Fourier Transform. The time-frequency duality plays a critical role in the selection of the sampling and truncation parameters and has to be addressed experimentally as a function of the inspected depth. To characterize a wide range of depths in a single test, the ideal solution is to sample at the maximum available frame rate for the longest possible time and to process all this data at once. Nevertheless, two factors restrict this operation. First, the maximum frame rate and storage capacity in any acquisition system are limited and so is the span of potentially detected depths. Second, although limited, the storage capacity generally exceeds the ordinary PC's capabilities to handle simultaneously all the collected information. As a result, a compromise between sampling and storage capacity, processing capabilities and range of potentially detected depth needs to be made. A four-step interactive methodology is proposed to deal with this problem. The idea is to perform a first partial processing with a fraction of the recorded data for visualization purposes only and then to individually manage selected (defective) areas, now visible, without repeating any test.