Research On Identification And Correction Of Internal And External Parameters Of 3D DIC System In Complex Environment

The research in the law of material deformation which is an important field of mechanics research is also of great importance to explore the law of material motion and constitutive relation in the matter.The modern optical image measurement technique,especially the DIC method,whose figure is active in mechanical,construction,materials,biology and other research fields is one of the most used displacement and deformation test methods in experimental mechanics,which provides significant test methods for exploring the law of matter motion and deformation.For another,with the development of science technology,the research object of mechanics is gradually expanding from traditional metallic materials,beam structure and general environment to hot non-metallic composite,variant cell topology and complex environment,which demands more displacement and deformation measurement technology to keep pace with the development of science technology.For this purpose,identification and calibration of internal and external parameters of threedimension DIC cameras image deformation testing system in a complex environment are mainly studied in this paper.The internal and external parameters of the three-dimensional DIC cameras image deformation testing system are the cornerstone and provide a guarantee for the accuracy of optical image measurement methods.However,under a complex environment,the camera paraments always will be changed because of the vibration,flow and heat et.al.Identifying and calibrating the variation of internal and external parameters in the system will be conducive to extending the scope and increasing the accuracy of optical image deformation measurement,which provides good data support for a study on mechanical properties of materials and structures in a complex environment.And considering the proposed influence,in the paper,the shock or vibration environment and the high-temperature environment are mainly researched.Based on the experimental data and statistical data,this paper studies the parameter acquisition,environmental impact,and calibration principle of a three-dimensional cameras system in sequence.In the optimization of parameter acquisition,the law of the calibration image sequence of the three-dimensional cameras system is analyzed.According to a large number of test calibration data in a complex environment,the optimal selection principle of calibration image sequence based on Markov chain model is established to optimize the parameter acquisition of cameras system.In terms of environmental impact,the change rule and evaluation method of cameras system parameters,aiming at the specific thermal environment and shock and vibration environment,are analyzed and constructed by combining optical interference experiments and random model simulation.In the aspect of principle and method for calibration,the semantic speckle with artificial characteristics other than deformation information added served for three-dimension cameras system identification and calibration in a complex environment is studied and designed according to the constructed parameter influence model and analysis of the characteristics of correlation speckle.By the technology of deep learning and image segmentation,the automatic identification and segmentation of semantic speckle region are realized,which lays a theoretical foundation for the intelligent operation of parameters of a three-dimension cameras system.Furthermore,based on the semantic speckle and the characteristics of fixed points in the environment,the acquisition and calibration of external parameters of the 3D DIC cameras test system are accomplished under the circumstance of large field impact vibration and the calibration method is proposed at the same time.Beyond that,the state space model of camera measurement in a thermal environment is also established.By combining the geometric and positioning constraints of semantic speckle,the camera projective space model,the matching algorithm and the Li operations,the fast parameter acquisition of the camera system in the large field of view and the rapid intelligent identification and real-time calibration of external parameters of three-dimension camera testing system under impact vibration environment are realized.The method not only overcomes the influence of the ambient light intensity change on image features but also solves the problem of instantaneous mutation of parameters caused by impact vibration.Meanwhile,aiming at the random disturbance of thermal flow in thermal environment,the test state space model of threedimension DIC cameras test system was built in line with the ideas of the equivalence principle and the characteristics of fixed points in the environment,which equates the random airflow disturbance to the camera internal parameter change.Also,in the constructed state space,fixed points feature and Kalman filtering principle are used to transform the complex random nonlinear process into linear observable state and real-time compensation for random thermal turbulence is realized.The method has achieved good results in large rotation camera pose reconstruction,strain measurement,and hot box test and has been successfully applied to impact failure test of large space steel structure and elastic modulus test of fiber-reinforced material under thermal environment.Both the experimental verification results and the engineering application test results show that the method can effectively identify and calibrate the internal and external parameters of a three-dimension camera deformation test system in a complex environment,which not only expands the application range of optical digital image measurement methods,improves the test precision,but also provides better data support for the study of mechanical properties of materials and structures in a complex environment,and has a promising application value.