Research On Key Techniques Of Phase-aided 3D Measurement System Calibration

As one of the most representative technologies in the field of optical three-dimensional(3D)measurement,phase-aided 3D measurement technology has been increasingly widely used in areas of advanced manufacturing,medical diagnosis,heritage protection and studio entertainment,due to the advantages of non-contact,high precision,high resolution,high universality,and rapid measurement.In the phase-aided 3D measurement technology,the 3D measurement precision is largely determined by the calibration accuracy,and how to improve the calibration accuracy of 3D measurement system is a key issue.This dissertation mainly focuses on the related methods of 3D measurement system calibration.The main research contents include three aspects as follows:(1)Circular marker is commonly used as the benchmark in the calibration,and its center localization accuracy is one of the important factors affecting the system calibration accuracy.Especially in the case of perspective projection transformation,center deviation errors are easily introduced.The method of recognizing and locating the circular markers is studied,and the error caused by perspective projection and lens distortion is analyzed.A high-precision center localization method is proposed.This method uses the calibration parameters to perform the back projection correction of the target images,and the center coordinates of the marks are extracted without perspective projection and lens distortion to recalibrate the camera iteratively,and finally the high-precision center coordinates of the markers and camera calibration parameters are obtained.The experimental results prove that the proposed method can effectively reduce the negative influence of perspective projection and lens distortion on the center localization of circular markers,and improve the camera calibration accuracy.(2)Since digital light processing(DLP)projector has been widely applied to encode structured light illumination,its optical model construction and parameters calibration are an important part of realizing 3D measurement.The optical path structure,imaging model and calibration method of the projection device are researched,and a novel method to accurately calibrate a DLP projector is proposed.The method maintains the optical axis of the camera perpendicular to the target plane to avoid the deviation of the center of the circle caused by the perspective projection of the camera imaging to the equivalent imaging of the projector,and uses the dual phase information of the projected structure-light to accurately obtain the corresponding relationship of different projection positions;at the same time,the bundle adjustment(BA)strategy is also added to optimize the 3D coordinates of the benchmarks and the projector calibration parameters.The experimental results demonstrate that the proposed method can accurately calibrate the parameters of a DLP projector.(3)In order to realize the automation of 3D measurement,a 3D measurement system based on a robot and a rotary table is built.The robot hand-eye calibration and rotary axis calibration of are studied to determine the transformation between the robot end-effector coordinate system and the 3D sensor coordinate system and the transformation between the rotation axis coordinate system and the 3D sensor coordinate system.And the improving algorithms are proposed respectively,which are verified both in simulations and experiments.And a system calibration solution was developed to integrate the robot and the rotary table to form a fully automated 3D measurement system,achieving multi-viewpoint and fully automatic 3D data acquisition and measurement by a single 3D sensor.