Research On Camera Calibration And Motion Parameter Measurement Technology In Large-Scale Videometrics

Camera measurement technology is widely used in motion parameter measurement tasks such as aerospace and national defense experiments.Through the calibration of internal and external parameters of optical measuring equipment such as camera,combined with target detection and tracking technology,the measurement of target motion parameters is realized to provide data support for target motion feature analysis.High precision calibration of camera internal and external parameters is not only the key technology of camera measurement,but also the basic premise of motion parameter measurement.However,the current field large-scale camera measurement application is faced with the problems of poor calibration environment and long system preparation time.Therefore,this paper mainly focuses on the application requirements of camera measurement in the field environment,and carries out the research on the internal parameter calibration of large field general camera,the external parameter self calibration of field large field binocular camera and the measurement of ground vehicle motion parameters.The main work of this paper is as follows:(1)Aiming at the cumbersome internal parameter calibration process of field large field camera,a general internal parameter calibration method of large field camera based on collimator is explored.The collimator with special star calibration pattern is customized to simulate the infinite calibration equipment,which can meet the imaging requirements of the camera equipped with telephoto lens.Then the collimator is photographed by the camera for many times,and the general camera calibration method is used to calibrate the internal parameters of the camera.This method can easily calibrate the camera internal parameters,and the calibration process is highly automated.It can be applied to the camera internal parameter calibration of longdistance focusing under field conditions.(2)Aiming at the difficulties of no scene information in the field of view in the external parameter calibration of field binocular camera,a self calibration method of external parameters of field binocular camera with large field of view based on UAV is proposed.Firstly,the UAV is used to create dynamic marker points in the public field of view of binocular camera,and the three-dimensional spatial coordinates of the marker points are not required.Aiming at the phenomenon of mismatching in the same name matching point pair of binocular camera,the graph cut optimization method is used to eliminate the outlier points,and the global optimization method is used to solve the optimal external parameters,so as to solve the problem that the nonlinear optimization algorithm is easy to converge to the local optimal solution.The robustness and accuracy of the method are verified by practical experiments.(3)According to the requirements of ground vehicle motion parameter measurement,the ground vehicle motion parameter measurement method based on UAV platform is studied.The motion parameter measurement method based on Lane corner extraction and the motion parameter measurement method based on homography matrix are studied and realized,and the measurement accuracy of ground vehicle motion parameters is further improved by the method based on ensemble learning.Practical experiments verify the effectiveness of the ground vehicle motion parameter measurement method based on UAV.Based on the vehicle speed measurement method,a vehicle motion parameter measurement software system based on UAV is designed and implemented.