A Calibration Method of 3D Laser Scanning Constrained by Point and Surface Features Simultaneously

Abstract

The invention relates to a three-dimensional laser scanning calibration method constrained by point and surface features at the same time, comprising the following steps: (1) Obtaining scanning observation values ​​and reference real values ​​of point and surface targets, and three-dimensional registration of two target point sets and gross error detection; (2) Use the registration conversion parameters as the initial pose parameters, and use the error model to describe the systematic deviation between the observed value and the reference value; (3) Set the system error parameters to zero and the initial pose parameters Perform least squares estimation to obtain optimized pose parameters and weights of observations; (4) Conditional adjustment with parameters constrained by points and surfaces to optimally estimate system error parameters and pose parameters; (5) Use the estimated parameters to correct the scanning observations, use mathematical statistics to correct the deviation between the observations and the reference values ​​before and after correction, and evaluate the accuracy and degree of improvement. Compared with the existing method, the present invention has the characteristics of synthesizing multiple constraints, simultaneously optimally estimating all parameters, good accuracy, high precision and easy operation.

Description

technical field [0001] The invention relates to a calibration method for a three-dimensional laser scanner, especially a three-dimensional laser scanner for dense point cloud imaging. At the same time, the high-precision coordinates of the reference point and the surface feature target are compared by comparing the deviation between the scanner observation value and the reference value. A calibration method for detecting, estimating, and correcting systematic errors in scanner observations. Background technique [0002] 3D laser scanning is a non-contact active measurement technology, which can quickly obtain accurate and dense 3D discrete point coordinates (point cloud data) on the surface of an object through pulsed or phased laser ranging and instantaneous deflection angle. copy". And according to the position and attitude of the scanner, the three-dimensional space coordinates in the external coordinate system are obtained through space coordinate transformation; it pro...

AI Technical Summary

Problems solved by technology

[0003] According to the shape characteristics of the reference target in the calibration process, the calibration methods that have appeared successively include: ①Surface-based calibration, which can conveniently select the reference datum plane on site and easily obtain a large number of redundant observations; but only provides one-dimensional Direction constraints, that is, the condition of the minimum distance from the point to the plane, lack of angle or orientation constraints, usually suitable for sparse point cloud situations

②Point-based calibration can effectively detect the deviation of the distance and the three-dimensional direction of two angles, but it needs to be evenly arranged and accurately measure a large number of point targets, which consumes manpower and material resources, and is generally suitable for dense point cloud situations

In addition, the calibration method of the 3D laser scanner constrained by point and surface features has not been reported yet, and the joint adjustment using multiple features is still rare

Images