The Study Of Method For Defect Detection In Terahertz Nondestructive Detection

The thesis proposes and investigates a new method based on terahertz technique to analyze internal defects of various samples. This method can obtain the internal defects states such as the sizes and distributions, through processing three-dimensional THz-wave scan data mathematically, which can be widely used.This method is described in three aspects in this thesis: theoretical research, experimental design and experimental results analysis. Among them, the Wavelet Transform theory and the Lipschitz Continuous Theory are the theoretical basis of this method. The time-frequency analysis of wavelet transform is suitable for extracting this kind of small mutation signal caused by defects, while the exponent tools of the Lipschitz Continuous Theory can well describe the changing speed and extent of the mutation signal. Therefore, it can obtain good results to measure the internal defects states of samples by using Lipschitz exponent on the basis of wavelet transform. In order to describe the defect state most intuitively, this thesis has designed a weight according to the statistical law of the Lipschitz exponent, to measure the possibility of a certain area within the sample to be a defect.This thesis designs a series of experiments to verify the validity of this method, and be strict in preparing sample, selecting instrument and designing algorithm. Multiple contrast experiments are conducted to confirm the important parameters of these experiments. The whole experiment is carried out under the principle of simulating the actual situation to ensure the representativeness of the experimental results.After some simple processing, the thesis makes a comprehensive analysis of the results. It shows that this method can obtain the internal defects states to a certain extent with high reliability. At the same time, it can freely adjust the severity of the judgment standards.