| As a high-precision,full-field measurement method,digital image correlation methods have been widely used in civil transportation,biomedicine,aerospace and other fields.The accuracy and consistency of the results of a measurement method are the most important,so it is necessary to study the measurement accuracy and consistency of the measurement results of digital image correlation methods.At present,there have been conclusions on the selection of speckle field parameters and the measurement system layout,but due to improper calibration and related calculation parameter selection,DIC measurement results are inconsistent and even measurement failure problems often occur.Therefore,it is necessary to determine appropriate calibration parameters and calculation parameters to ensure the accuracy and consistency of DIC measurement results.This dissertation first studies the effect of calibration parameters on the measurement results of digital image correlation methods.Starting from the recognition accuracy of the characteristic points of the calibration board,a sub-pixel translation simulation test was designed based on the C ++platform,and the robustness and anti-noise ability of the circle center recognition algorithm of the circular array calibration board under dynamics was obtained.The conclusion of the algorithm.In the case of noise,the recognition accuracy of the two is almost the same,and the maximum recognition error is 0.365 pixels.Through the simulation projection experiment,the influence of the accuracy of the feature point recognition of the calibration plate on the camera calibration results is discussed,and the internal and external parameters and the three directions of x,y,and z are analyzed respectively.Based on this,the influence of feature point recognition accuracy on 3D reconstruction is studied.It is found that the 3D reconstruction error is linearly correlated with the feature point recognition accuracy,and the slope of the trend line is less than 1.In the study on the influence of calculation parameters on the measurement results of digital image correlation methods,first,the effects of each calculation parameter on the DIC measurement results are briefly described in theory,and then the displacement measurement results are obtained from the translation test and the aluminum alloy plate tensile test respectively With regard to both the strain measurement results and the strain measurement results,the influence of the template size and calculation step size on the DIC measurement results was studied.The related calculation results show that when the template size is 29×29pixel,the value of the calculation step within the range of3-17 pixel has little effect on the DIC measurement result.The fixed calculation step is 7pixel,and the template side length value is 11-39 pixel.It has little effect on the DIC measurement displacement results;when the template side length is less than 11 pixel,it will seriously affect the DIC strain measurement result.The consistency of DIC displacement measurement results and strain measurement results was verified by 20 sets of glass plate translation experiments and 20 sets of printed speckle aluminum alloy standard plate tensile tests.Using the coefficient of variation as the consistency evaluation index,the results show that the coefficient of variation of the 20-time translational DIC measurement result is 0.25%,and the coefficients of variation of the elastic modulus and Poisson's ratio data measured by 20 times of tensile tests are 0.399% and 1.424%,Both are less than 5% required by the coefficient of variation.It is proved that the displacement measurement results and strain measurement results of the digital image correlation method are consistent.Finally,based on the accuracy and consistency of the DIC measurement results,the DIC measurement method is applied to the actual project,and the full-field deformation measurement of the granular material compression process is completed. |
PDF Full Text Request