| Research on linear structured light measurement technology has rapidly progressed in recent years.Since handheld scanning equipment does not need to rely on precision moving mechanical transmission structures during the measurement process,the research on it has gradually increased.At the same time,in order to eliminate the installation error caused by the disassembly and measurement of parts during processing,in-place measurement is essential.The hand-held linear structured light scanning and measuring system can measure the clamped workpiece on the machine tool in a flexible position,and realize the guided processing of the machine tool,making the processing more accurate and convenient.This paper focuses on the construction of a handheld linear structured light scanning measurement system,the optimization of handheld stereo target position and orientation estimation,the extraction of the ellipse feature point center,the screening of the hand-eye calibration image sequence,the correction of the system model,the restoration of scanning surface shape and the secondary processing of guided machine tool.The main work and innovation points are as follows:1)This paper studies the visual calibration method analyzes the state of moving objects under the camera view,and designs a feature point target with a unique coding sequence based on the structure of handheld mobile devices.The target feature marker points are designed as asymmetric and non coplanar three-dimensional distributions,avoiding the influence of pose ambiguity on the spatial pose estimation of handheld mobile devices.2)An algorithm for optimizing the spatial coordinate transformation matrix based on the Kalman filter is studied.The matrix is decomposed into 6 elements and filtered separately,and then recombined to improve the accuracy of the spatial coordinate transformation matrix.From the accuracy,orthogonality,and robustness of the algorithm,the results before and after the calculation and optimization of external parameters are compared and analyzed,and the theoretical derivation,simulation analysis,and actual motion experiment verification are carried out for the algorithm.The scanning verification of high-precision planes and milling workpieces shows that the algorithm improves the accuracy of handheld scanning and measuring equipment.3)The whole system is analyzed from the angle of error model.In the aspect of the hardware model,the influence of the angle between the laser plane and the mobile camera on the scanning measurement results is analyzed for the research of the handheld device mentioned in this paper,and the optimal model is obtained.Building the mathematical model of the whole measurement system,analyzing the deviation between the measured point cloud and the ideal point cloud,and obtaining the error compensation correction matrix,so as to realize the error compensation of the whole measurement system model and improve the equipment measurement accuracy.4)Aiming at the problem of inaccurate hand-eye calibration matrix in the calibration process of handheld mobile devices,a simple and convenient high-quality image sequence filtering method is proposed.The obtained initial image sequence is arranged and combined,and the calibration matrix is calculated separately,and then all the matrix results are thresholded to eliminate the image sequence with gross error,and finally,the optimal initial image sequence is obtained.This method reduces or avoids complex iterative calculation,and effectively improves the efficiency and accuracy of hand-eye calibration matrix calculation of handheld mobile device module.5)An edge fitting algorithm based on the change of ellipse gray gradient is proposed to solve the problem of inaccurate feature point center extraction caused by ellipse imaging in the camera.The ellipse in the image is segmented,and the X and Y directions are fused respectively to obtain more accurate ellipse edge points.Ellipse fitting and LM algorithm optimization are performed on the edge points to obtain the feature point center.The algorithm can improve the recognition accuracy of feature points and meet real-time requirements,making the real-time positioning of handheld mobile devices more accurate.6)The measurement results of the hand-held linear structured light scanning and measuring equipment interact with the machine tool,and the tool guidance of the machine tool is realized by scanning and measuring the clamping workpiece on the machine tool.The scanning results of the in-place measurement of the workpiece are reconstructed according to the position of the tool,and then the tool path is planned through the measurement results to guide the tool to realize the secondary processing.7)Measuring the height of the step surface and the free surface,and the system measurement accuracy.The experiment shows that the maximum error of the step section measurement of the handheld linear structured light scanning measurement system is less than 0.03 mm,which can well realize the free surface measurement and surface restoration,and the system measurement accuracy is less than 0.1mm. |
PDF Full Text Request