Model-based parameter estimation for defect characterization in ultrasonic NDE applications

In this study, a model-based parameter estimation algorithm is applied to characterize defects for ultrasonic NDE applications. Chirplet, a signal model representing a broad range of ultrasound echoes, is utilized for signal decomposition and parameter estimation. The model-based parameter estimation algorithm and its results are discussed in the context of system modeling. Through simulation and benchmark data analysis, the effectiveness of model-based signal processing has been successfully demonstrated. This method can be used as a quantitative method for applications in flaw detection, defect estimation, pattern recognition and material characterization.