Optimization And Experimental Study Of Image Correlation Algorithm

Digital image correlation in experimental mechanics has a pivotal position and a very wide range of applications, especially in the field of optical measurement of mechanical measuring surface displacement and strain fields play a very important role. With the recent decades the rapid development of manufacturing industry, the demand for all kinds of metal materials are increasing. Digital image correlation method as the experimental measurement of mechanical properties along with the mechanical means has been applied so as the optical, electrical and other fields.Currently, the domestic digital image correlation method of research in two directions, the accuracy and efficiency. On the accuracy of the research focuses on the various sub-pixel location method; and the problems are rarely involved in computational efficiency. In this paper, after a series of simulation and experimental studies, found that the current in the calculation of the efficiency of computational algorithms related to the existence of a huge room for improvement. Therefore, this part from the whole set in pixels, a new integer pixel search method - Fast search algorithm for storage, to some extent improve the computational efficiency of the relevant calculation. The accuracy of research, this paper studied the two most commonly used sub-pixel calculation method: gradient and surface fitting. And from the computational accuracy, efficiency and stability of the comparison and study other aspects of the two sub-pixel algorithm, came to the conclusion from the results. Upon completion of the research work of the algorithm, this article was standard tensile uniaxial tensile test specimens, and the hole uniaxial tensile test specimens will be calculated by the displacement field associated with the simulated displacement field comparison, were obtained satisfactory results, which verifies the accuracy of the algorithm and its stability. Finally, this research work done to make a detailed summary, and digital image correlation method in the future development of recommendations and forecasts.