Research On Calibration Of External Parameters Of Binocular Cloud Height System Based On Odometer Method

Cloud is an important weather phenomenon,which plays an important role in geophysical processes such as climate change and atmospheric radiation balance.The current research on clouds mainly focuses on the study of cloud volume,cloud shape,and cloud base height parameters.This paper mainly focuses on the related research on the external parameter calibration of the ground-based binocular cloud height observation system,and elaborates on the construction of the binocular cloud height system,camera calibration,and experiment in three parts.This article first introduces the self-built ground-based binocular vision cloud height observation system,briefly describes the system hardware structure and software structure,and gives a detailed introduction to the imaging system and data analysis and processing modules.Then the selection of the focal length and baseline of the system camera was determined through the principle and process of binocular ranging,and finally an outdoor binocular cloud height observation system was built to collect the original sky cloud image.Camera calibration plays an important role in the binocular vision measurement of cloud base height.The accuracy of the calibration data directly affects the detection accuracy of cloud base height.The binocular camera has a long baseline.In view of the difficulty of the traditional camera calibration method and the low calibration accuracy,this paper proposes a odometer-based external parameter calibration method of the binocular camera,using the improved feature point extraction and matching algorithm based on SURF Obtain the point pairs of the left and right cloud images with the same name,and use the RANSAC algorithm to filter and eliminate the mismatched point pairs generated.Then,according to the epipolar geometry constraint theorem,the constraint conditions containing the rotation and translation matrix between the left and right cameras are obtained,and the external parameters of the camera are solved according to the constraint conditions.In order to verify the accuracy of the odometer method proposed in this paper,the feature point extraction and matching experiments were carried out.The algorithm in this paper was compared with the classic SIFT and SURF feature point extraction and matching algorithms.Through the analysis of the experimental results,it was concluded that the accuracy of the algorithm in this paper was better.High,better real-time performance.In order to verify the feasibility and accuracy of the odometer method used in camera external parameter calibration,this article carried out the indoor small baseline external parameter calibration experiment and outdoor large baseline calibration experiment respectively.First,build a binocular ranging platform with a small baseline of 30 mm in the room,and use the odometer-based camera calibration method and the dual target calibration method to calibrate the camera's external parameters,and calibrate the horizontal distance of the binocular camera based on the odometer external parameter calibration method.It is 300.1323 mm,and the error is only 0.1323 mm,which is more accurate than the dual-objective method.Then,the comparative analysis of the binocular correction map and the disparity map obtained according to the external parameter data further verified the accuracy of calibrating the external parameters of the binocular camera by the odometry method.Then set up a binocular cloud height system with a baseline of60 meters outdoors,and use the odometry-based binocular camera external parameter calibration method.The distance between the two cameras is 60.3444 meters,the error is only 34.44 cm,and the accuracy is up to 99.43%,and by analyzing the corrected binocular cloud image and sky disparity map,the validity of the odometry method to calibrate the external parameters of the camera is further verified,and the feasibility of the odometry method for the external parameter calibration of the binocular cloud height detection system.Through the research and work of the above three parts,this subject has completed the research on the external parameter calibration of the binocular cloud height system based on the odometry method,and solved the problems of the external parameter calibration of the binocular cloud height system and the inaccurate calibration.The content of the research It has important practical significance and provides researchers with new research ideas in Yungao inversion work.