A Study On The Application Of Image Transform In Digital Speckle Correlation Method

In this thesis, the main concerned problems are to improve the correlation search in the digital speckle correlation method (DSCM) and to design an integrated software system of correlation measurement. Nowadays, two development trends in practical applications of the DSCM are to increase the convergent speed and improve the precision. For this purpose, a novel coarse-fine search method (positon predetermination method) is firstly developed. This approach is based on affine transform and a new technique of fine search called "nested fine search method". The features of the new method are as follows: firstly, the nested fine search method is proved to be having the similar accuracy with but higher calculating speed than the previous fine search scheme; second, as the approximate rigid displacement and the displacements cause by deformation can be directly and automatically calculated by affine transform, surpassing the rigid displacement of each sample pixel, a relatively smaller searching region size, which might decrease the probability of miss-identification and raise the quality of identification, and less calculating times for coarse search are achieved. As a result, the novel scheme becomes more automated and more intelligent than the previous ones. Furthermore, the affine transform is also used to set the initial value of the parameter vector, which includes the information of displacement vector and its derivatives, of the iteration method, resulting in several new approaches of initial value setting. Comparison of the new methods with the old ones by simulated images revealed that higher efficient and accuracy could be achieved by introducing affine transform into initial parameter vector. Subsequently, an integrated and easily-used software with muti-functions and friendly Graphic User Interface (GUI) is developed based on the previous techniquesand the above mentioned positon predetermination method and initial value seting approach. This software provides several approaches of searching region setting and functions such as displacement and strain measurement, poission ratio measurement and simulated image generation, etc. At last, investigation of size effect on the fracture behavior of low-dimensional material has been carried out base on the work of Doctor Huaiwen Wang [131]. Firstly, principle and experimental technique of measuring fracture toughness JC by DSCM is introdced. Second, fracture toughness JC of copper foils with thicknesses t ranging from 0.02 to 1mm were evaluated systematically and the fracture toughness vs. thickness curve was obtained quantificationally. Finally, the influce of thickness on fracture toughness is especially discussed associated with the work required per unit area during crack tip necking, the work required per unit area during material separation, micro fracture feature and stress triaxiality.