CN105203023A

CN105203023A - One-stop calibration method for arrangement parameters of vehicle-mounted three-dimensional laser scanning system

Info

Publication number: CN105203023A
Application number: CN201510406380.3A
Authority: CN
Inventors: 王力; 李广云; 李明磊; 李宗春; 周阳林; 杨振; 邓勇; 刘灵杰; 宗文鹏; 彭逸凡
Original assignee: PLA Information Engineering University
Current assignee: PLA Information Engineering University
Priority date: 2015-07-10
Filing date: 2015-07-10
Publication date: 2015-12-30
Anticipated expiration: 2035-07-10
Also published as: CN105203023B

Abstract

The invention relates to a one-stop calibration method for setting parameters of a vehicle-mounted three-dimensional laser scanning system. By fixing a certain number of artificial markers with reflective properties on objects in the calibration field, the geocentric coordinates of these artificial marker points are obtained. Select a certain location with marker points as the site, let the vehicle equipped with the 3D laser scanning system stop at the site, use the 3D acquisition method to collect a series of marker point clouds, identify and locate these marker point clouds, and establish In the laser scanning coordinate system L. The geocentric coordinates of artificial signs are converted from the geocentric coordinate system (WGS84) to the inertial platform coordinate system (I) through coordinate transformation, and the coordinates in the laser scanning coordinate system (L) are transformed through the coordinate transformation model to obtain the placement parameter. The method can be calibrated at only one station, reduces the complexity of the mathematical conversion model, and improves the efficiency of calibration of placement parameters.

Description

A one-stop calibration method for the placement parameters of a vehicle-mounted 3D laser scanning system

技术领域 technical field

本发明涉及一种车载三维激光扫描系统安置参数的一站式标定方法。属于利用光学为特征的测量技术领域。 The invention relates to a one-stop calibration method for setting parameters of a vehicle-mounted three-dimensional laser scanning system. The invention belongs to the field of measuring technology characterized by optics.

背景技术 Background technique

随着激光扫描速度、精度以及海量数据处理能力的不断提高，激光扫描仪逐步成为车载移动测量系统的主要传感器，称为车载激光扫描系统。 With the continuous improvement of laser scanning speed, accuracy and massive data processing capabilities, laser scanners have gradually become the main sensor of vehicle-mounted mobile measurement systems, called vehicle-mounted laser scanning systems.

目前车载三维激光扫描系统安置参数标定的方法主要是通过以二维采集方式设置不同的水平扫描角，需要让车辆在行驶中进行按照一定的扫描角度进行测量，对测量物体的平面度要求较高；而且需要测量多次，获得不同扫描角度下的测量数据，即二维扫描方式，其过程较为繁琐。然后，将扫描后的目标点和标志点地心坐标直接建立整体的安置参数转换模型，根据公共点的地心坐标和瞬时点云坐标，通过最小二乘法估计参数，涉及了多次转换，模型复杂度高，而且两套坐标尺度差距较大，直接计算容易产生舍入误差。 At present, the method of calibrating the installation parameters of the vehicle-mounted 3D laser scanning system is mainly to set different horizontal scanning angles by means of two-dimensional acquisition. It is necessary to allow the vehicle to measure according to a certain scanning angle while driving, which requires high flatness of the measured object. ; and it needs to be measured multiple times to obtain measurement data at different scanning angles, that is, two-dimensional scanning, and the process is relatively cumbersome. Then, the geocentric coordinates of the scanned target points and marker points are used to directly establish the overall placement parameter conversion model, and the parameters are estimated by the least square method according to the geocentric coordinates of the public points and the instantaneous point cloud coordinates, which involves multiple transformations. The complexity is high, and the difference between the two sets of coordinate scales is large, and direct calculation is prone to rounding errors.

发明内容 Contents of the invention

本发明目的在于克服现有技术的不足，提出了一种对车载三维激光扫描系统进行一站式的标定方法，用于解决现有技术标定方法复杂导致的标定效率低的问题。 The purpose of the present invention is to overcome the shortcomings of the prior art, and propose a one-stop calibration method for a vehicle-mounted three-dimensional laser scanning system, which is used to solve the problem of low calibration efficiency caused by the complexity of the calibration method in the prior art.

本发明的上述目的主要是通过如下技术方案予以实现的： Above-mentioned purpose of the present invention is mainly achieved through the following technical solutions:

一种车载三维激光扫描系统安置参数的一站式标定方法，包括如下步骤： A one-stop calibration method for setting parameters of a vehicle-mounted three-dimensional laser scanning system, comprising the following steps:

步骤(一)、在标定场中的物体上固定一定数量的具有反光特性的人工标志，获取这些人工标志点的地心坐标； Step (1), fixing a certain number of artificial markers with reflective properties on the objects in the calibration field, and obtaining the geocentric coordinates of these artificial marker points;

步骤(二)、选取标志点的某处位置作为站点，让载有三维激光扫描系统的车辆停止在该站点，利用激光扫描系统对物体进行三维扫描，采集得到一系列标志点云。 Step (2), select a certain location of the marker point as a station, let the vehicle carrying the 3D laser scanning system stop at the station, use the laser scanning system to scan the object in 3D, and collect a series of marker point clouds.

步骤(三)、对这些标志点云进行识别与定位，获取其在激光扫描坐标系L中的坐标。 Step (3), identifying and locating these mark point clouds, and obtaining their coordinates in the laser scanning coordinate system L.

步骤(四)、人工标志的地心坐标通过坐标转换方式从地心坐标系(WGS84)转换到惯性平台坐标系(I)，并和激光扫描坐标系(L)中的坐标通过坐标转换模型实现变换，求得安置参数。 Step (4), the geocentric coordinates of the artificial marks are converted from the geocentric coordinate system (WGS84) to the inertial platform coordinate system (I) through coordinate transformation, and the coordinates in the laser scanning coordinate system (L) are realized through the coordinate transformation model Transform to obtain the placement parameters.

进一步的，步骤(三)所述的对标志点云进行识别与定位方法是：1)通过对标志点云进行阈值分割，2)将落在接收阈内的标志点云聚类，3)剔除噪声目标，4)确定标志点在激光扫描坐标系的重心坐标。 Further, the method for identifying and locating the landmark point cloud described in step (3) is: 1) performing threshold segmentation on the landmark point cloud, 2) clustering the landmark point cloud falling within the receiving threshold, and 3) eliminating Noise target, 4) Determine the barycentric coordinates of the marker point in the laser scanning coordinate system.

进一步的，所述阈值分割过程中，对于落在接受阈内的标志点给予保留，阈值随标志点的回光反射距离和角度不同而变化。 Further, in the threshold segmentation process, mark points falling within the acceptance threshold are reserved, and the threshold value varies with the reflective distance and angle of the mark points.

进一步的，设置标志点初始重心，标志点云采用逐一分类的方式，计算标志点重心，对重心进行重建实现自增长聚类。 Further, the initial center of gravity of the landmarks is set, the landmark point cloud is classified one by one, the center of gravity of the landmarks is calculated, and the center of gravity is reconstructed to achieve self-growth clustering.

本发明与现有技术相比的有益效果是： The beneficial effect of the present invention compared with prior art is:

以往利用三维激光扫描系统标定参数是通过以二维采集方式设置一定的水平扫描角度多次测量，利用载有激光扫描系统的车辆在行驶中测量的结果进行参数解算，过程较为复杂，效率低，影响测量的数据精度。该发明在保证定姿定位系统正常工作后，在标定场中采用让载有激光扫描系统的车辆短暂停止，并以三维采集方式，迅速采集标志的点云数据。因为仅在一站进行测量，大大提高了标定的效率。进而，由于获得的数据与采用二维扫描方式不同，所以在建立安置参数解算数学模型时，可以将公共点的绝对坐标先通过定姿定位结果转换到载体坐标系，再和激光扫描坐标系的数据结合求解安置参数，从而将复杂的参数估计数学模型简化成单次坐标转换的参数解算模型，解算效率高。 In the past, the calibration parameters of the 3D laser scanning system were measured multiple times by setting a certain horizontal scanning angle in a 2D acquisition method, and the parameters were calculated by using the measurement results of the vehicle equipped with the laser scanning system while driving. The process is relatively complicated and the efficiency is low. , affecting the accuracy of the measured data. After ensuring the normal operation of the attitude determination and positioning system, the invention temporarily stops the vehicle carrying the laser scanning system in the calibration field, and quickly collects the point cloud data of the mark in a three-dimensional collection method. Because the measurement is performed at only one station, the efficiency of calibration is greatly improved. Furthermore, since the obtained data is different from the two-dimensional scanning method, when establishing the mathematical model for solving the placement parameters, the absolute coordinates of the public points can be converted to the carrier coordinate system through the attitude positioning results first, and then combined with the laser scanning coordinate system Combined with the data to solve the placement parameters, the complex parameter estimation mathematical model is simplified into a single coordinate conversion parameter calculation model, and the calculation efficiency is high.

而且，利用激光扫描系统采集的标志点云，通过移动阈值分割、自增长聚类分析等算法实现对人工标志的自动定位和识别，也能相应提高精度。 Moreover, using the point cloud of signs collected by the laser scanning system, the automatic positioning and recognition of artificial signs can be realized through algorithms such as moving threshold segmentation and self-increasing cluster analysis, which can also improve the accuracy accordingly.

附图说明 Description of drawings

图1是本发明安置参数标定场； Fig. 1 is the calibration field of placement parameters of the present invention;

图2是本发明标定场中人工标志的布设； Fig. 2 is the layout of artificial signs in the calibration field of the present invention;

图3是本发明安置参数解算模型的流程图。 Fig. 3 is a flow chart of the placement parameter calculation model of the present invention.

具体实施方式 Detailed ways

下面结合附图对本发明做进一步详细的说明。 The present invention will be described in further detail below in conjunction with the accompanying drawings.

(一)、人工标志的地心坐标获取 (1) Acquisition of geocentric coordinates of artificial signs

如图1所示为本发明安置参数标定场，在标定场选取已知点S₁，并且已知该点的高精度坐标，在检定场另选5个控制点，S₅和S₆是两个可以相互通视的控制点。 As shown in Figure 1, the parameter calibration field of the present invention is arranged. A known point S ₁ is selected in the calibration field, and the high-precision coordinates of this point are known, and 5 control points are selected in the verification field. S ₅ and S ₆ are two A control point that can communicate with each other.

如图2所示为本发明标定场中人工标志的布设，在S₅附近的物体上固定一定数量的具有反光特性的人工标志。在S₅设置全站仪，S₆架设棱镜，利用方向观测法测角和激光测距获取人工标志的站心坐标。设站心坐标系的原点A的坐标为(B₀,L₀,H₀)，则对应的的空间直角坐标系形式为(X₀,Y₀,Z₀)。任一一点地心坐标(X,Y,Z)与站心坐标(N,E,U)之间的转换关系，如下式表示： As shown in Figure 2, the arrangement of artificial signs in the calibration field of the present invention, a certain number of artificial signs with reflective properties are fixed _on objects near S5. A total station is set up at _S5 , a prism is set up at _S6 , and the center coordinates of artificial signs are obtained by using the direction observation method to measure angles and laser ranging. Assuming that the coordinates of the origin A of the station center coordinate system are (B ₀ , L ₀ , H ₀ ), the corresponding space rectangular coordinate system is in the form of (X ₀ , Y ₀ , Z ₀ ). The conversion relationship between any point geocentric coordinates (X, Y, Z) and station center coordinates (N, E, U) is expressed in the following formula:

$[\begin{matrix} N N \\ E E. \\ U u \end{matrix}] = = T T \cdot \cdot (([\begin{matrix} X x \\ Y Y \\ Z Z \end{matrix}] - - [\begin{matrix} {X x}_{00} \\ {Y Y}_{00} \\ {Z Z}_{00} \end{matrix}])) - - - - - - ((11))$

式中： In the formula:

$T T = = [\begin{matrix} - - {sinB sinB}_{00} {cosL cos L}_{00} & - - {sinB sinB}_{00} {sinL sin L}_{00} & {cosB cosB}_{00} \\ - - {sinL sin L}_{00} & {cosL cos L}_{00} & 00 \\ {cosB cosB}_{00} {cosL cos L}_{00} & {cosB cosB}_{00} {sinL sin L}_{00} & {sinB sinB}_{00} \end{matrix}] - - - - - - ((22))$

通过上式的逆变换就可以得到人工标志的地心坐标。 Through the inverse transformation of the above formula, the geocentric coordinates of the artificial signs can be obtained.

(二)、利用激光扫描系统采集三维标志点云 (2) Using laser scanning system to collect 3D mark point cloud

在标定场中选取一处测量站点，驾驶载有激光扫描系统的车辆行至该站点并停止，利用车载三维激光扫描系统通过三维采集方式，对安放人工标志的物体进行扫描，采集得到一系列标志点云。 Select a measurement site in the calibration field, drive a vehicle equipped with a laser scanning system to the site and stop, use the vehicle-mounted 3D laser scanning system to scan the object with artificial marks through the 3D acquisition method, and collect a series of signs point cloud.

(三)、标志点云的自动识别与定位 (3) Automatic recognition and positioning of landmark point clouds

1、利用激光扫描仪得到采集点的距离信息，通过KNN计算每个点的邻阈并拟合成平面，其中目标点与一起中心的连线与平面法线间的夹角为入射角，将距离和入射角代入整体模型，得到回光反射标志在该条件下期望的回光反射值(即为阈值)，对于落在该接收阈内的点给予保留。 1. Use the laser scanner to obtain the distance information of the collection point, calculate the adjacent threshold of each point through KNN and fit it into a plane, where the angle between the line connecting the target point and the center of the plane and the plane normal is the incident angle, and the The distance and angle of incidence are substituted into the overall model to obtain the expected retroreflective value (threshold) of the retroreflective sign under this condition, and the points falling within the acceptance threshold are reserved.

2、将测量目标组成集合T，每个目标由激光系统采集的多个激光点组成。对每个目标定义一个定位点O，用单个目标所包含点的重心来表示。若目标T_i由n个点组成，则T_i的重心O为： 2. The measurement targets are formed into a set T, and each target is composed of multiple laser points collected by the laser system. Define an anchor point O for each target, represented by the center of gravity of the points contained in a single target. If the target T _i is composed of n points, then the center of gravity O of T _i is:

$\begin{matrix} {O o}_{x x} = = \frac{{Σ Σ}_{i i = = 11}^{n no} {x x}_{i i}}{n no} & {O o}_{y the y} = = \frac{{Σ Σ}_{i i = = 11}^{n no} {y the y}_{i i}}{n no} & {O o}_{z z} = = \frac{{Σ Σ}_{i i = = 11}^{n no} {z z}_{i i}}{n no} \end{matrix} - - - - - - ((33))$

从分割中得到的点集中任选一点作为第一目标点T_i，目标的重心坐标用该点的坐标进行初始化。遍历分割后的点集，对分割后的点集进行编号，按照顺序比较目标点和目标重心O的欧式距离l；对于第i点，若l小与目标直径d，则i点属于该目标，将其加入该目标，并更新目标的重心；若第i点不属于任何目标，则重新建立一个新的目标，其重心初始化为该点的坐标。对于不同类别目标用不同颜色表示出来。 A point is selected from the point set obtained in the segmentation as the first target point T _i , and the coordinates of the center of gravity of the target are initialized with the coordinates of this point. Traverse the divided point set, number the divided point set, compare the Euclidean distance l between the target point and the target center of gravity O in order; for the i-th point, if l is smaller than the target diameter d, then the i point belongs to the target, Add it to the target, and update the center of gravity of the target; if the i-th point does not belong to any target, re-establish a new target, and its center of gravity is initialized to the coordinates of this point. Different colors are used for different categories of objects.

3、在激光采集系统扫描得到的点集中会存在噪声点，利用平面度、尺寸、期望的点数等点集的特性，对噪声点进行剔除。 3. There will be noise points in the point set scanned by the laser acquisition system. Use the characteristics of the point set such as flatness, size, and expected number of points to eliminate the noise points.

4、对于反射标志，可以采用回光强度加权的方法求取重心坐标，如下式所示： 4. For reflective signs, the weighted method of return light intensity can be used to obtain the coordinates of the center of gravity, as shown in the following formula:

$\begin{matrix} {x x}_{T T i i} = = \frac{{Σ Σ}_{j j = = 11}^{K K} {A A}_{j j} {x x}_{j j}}{K K} & {y the y}_{T T i i} = = \frac{{Σ Σ}_{j j = = 11}^{K K} {A A}_{j j} {y the y}_{j j}}{K K} & {z z}_{T T i i} = = \frac{{Σ Σ}_{j j = = 11}^{K K} {A A}_{j j} {z z}_{j j}}{K K} \end{matrix} - - - - - - ((44))$

其中：K为第i个目标T_i的点数，A_j为第j点的回光强度。 Among them: K is the point number of the i-th target T _i , and A _j is the return light intensity of the j-th point.

将得到的标志点坐标建立在激光扫描坐标系L中。 The coordinates of the obtained marker points are established in the laser scanning coordinate system L.

以上聚类实施方式中，采用了欧式距离法实现目标点的自增长聚类，也可以采用k-均值聚类算法，最邻近聚类算法等。 In the above clustering implementation, the Euclidean distance method is used to realize the self-growth clustering of the target points, and the k-means clustering algorithm, the nearest neighbor clustering algorithm, etc. can also be used.

(四)、建立安置参数解算模型 (4) Establish a resettlement parameter calculation model

如图3所示为本发明安置参数解算模型的流程图，人工标志的地心坐标通过坐标转换方式转换到载体坐标系(惯性平台坐标系)，具体方式如下： As shown in Figure 3, it is a flow chart of the placement parameter solution model of the present invention, the geocentric coordinates of the artificial signs are converted to the carrier coordinate system (inertial platform coordinate system) by a coordinate conversion mode, and the specific methods are as follows:

(1)地心坐标系WGS84转换到当地水平坐标系LH (1) Convert the geocentric coordinate system WGS84 to the local horizontal coordinate system LH

车载激光扫描系统中主GNSS接收机相位中心坐标为(B,L,H)，直角坐标形式为设工人标志坐标为(x₈₄,y₈₄,z₈₄)^T。将WGS84坐标系绕Z轴逆时针旋转L，在绕Y轴顺时针旋转90度+B，得到旋转矩阵R_W，可完成旋转变换。综合得到标志中心坐标从WGS84到LH的变换模型，如下式： The phase center coordinates of the main GNSS receiver in the vehicle-mounted laser scanning system are (B, L, H), and the rectangular coordinates are in the form Let the coordinates of the worker's logo be (x ₈₄ , y ₈₄ , z ₈₄ ) ^T . Rotate the WGS84 coordinate system counterclockwise L around the Z axis, and rotate clockwise 90 degrees + B around the Y axis to obtain the rotation matrix R _W , which can complete the rotation transformation. The transformation model of the logo center coordinates from WGS84 to LH is obtained comprehensively, as follows:

$[\begin{matrix} {x x}_{L L H h} \\ {y the y}_{L L H h} \\ {z z}_{L L H h} \end{matrix}] = = {R R}_{W W} [\begin{matrix} {x x}_{8484} - - {x x}_{8484}^{G G} \\ {y the y}_{8484} - - {y the y}_{8484}^{G G} \\ {z z}_{8484} - - {z z}_{8484}^{G G} \end{matrix}] - - - - - - ((55))$

(2)当地水平坐标系LH转换到惯性平台坐标系I (2) Transform the local horizontal coordinate system LH to the inertial platform coordinate system I

惯性平台坐标系I中存放的是载体的定姿定位坐标信息，利用定姿定位系统的三个瞬时姿态角(航向角、侧滚角、仰俯角)确定由LH到I的三个旋转欧拉角，继而得到旋转矩阵R_N。惯导元件在进行导航时，加入了重力异常的补偿，使得三个欧拉角是以当地水平参考系(以参考椭球的法线作为Z轴)为参考坐标系，因此，垂线偏差得到了改正。将I的坐标原点定位天线相位中心，人工标志在I系下的坐标为(x_I,y_I,z_I)^T，则标志地心坐标到惯性平台坐标的转换模型如下式： The attitude determination and positioning coordinate information of the carrier is stored in the inertial platform coordinate system I, and the three instantaneous attitude angles (heading angle, roll angle, and pitch angle) of the attitude determination and positioning system are used to determine the three rotation Euler angles from LH to I Angle, and then get the rotation matrix R _N . When the inertial navigation element is navigating, compensation for gravity anomalies is added, so that the three Euler angles are based on the local horizontal reference system (with the normal of the reference ellipsoid as the Z axis) as the reference coordinate system. Therefore, the vertical line deviation is obtained Corrected. The origin of the coordinates of I is positioned at the antenna phase center, and the coordinates of the artificial marker in the I system are (x _I , y _I , z _I ) ^T , then the conversion model from the geocentric coordinates of the marker to the coordinates of the inertial platform is as follows:

$[\begin{matrix} {x x}_{I I} \\ {y the y}_{I I} \\ {z z}_{I I} \end{matrix}] = = {R R}_{N N} [\begin{matrix} {x x}_{L L H h} \\ {y the y}_{L L H h} \\ {z z}_{L L H h} \end{matrix}] - - - - - - ((66))$

(3)安置参数解算 (3) Calculation of placement parameters

激光坐标系和惯性平台坐标系都为三维右手直角坐标系，且两者间为刚性变换，通过3个平移参数和3个旋转参数(α,β,γ)建立转换模型进行转换，坐标转换模型如下式： Both the laser coordinate system and the inertial platform coordinate system are three-dimensional right-handed Cartesian coordinate systems, and the two are rigid transformations, through three translation parameters Establish a conversion model with three rotation parameters (α, β, γ) for conversion. The coordinate conversion model is as follows:

$[\begin{matrix} {x x}_{I I} \\ {y the y}_{I I} \\ {z z}_{I I} \end{matrix}] = = {R R}_{M m} [\begin{matrix} {x x}_{L L} \\ {y the y}_{L L} \\ {z z}_{L L} \end{matrix}] + + [\begin{matrix} Δ Δ {x x}_{L L}^{I I} \\ {Δy Δy}_{L L}^{I I} \\ {Δz Δz}_{L L}^{I I} \end{matrix}] - - - - - - ((77))$

其中，[x_L，y_L，z_L]^T为激光坐标系L中的坐标； Wherein, [x _L , y _L , z _L ] ^T is the coordinate in the laser coordinate system L;

通过转换模型可以计算出平移参数和旋转参数(即为6个安置参数)。其中，旋转参数从旋转矩阵R_M求得，其中： The translation parameters and rotation parameters (that is, 6 placement parameters) can be calculated by converting the model. Among them, the rotation parameters are obtained from the rotation matrix R _M , where:

R_M＝R(γ)R(β)R(α)(8) R _M =R(γ)R(β)R(α)(8)

$R R ((γ γ)) = = [\begin{matrix} c c o o s the s γ γ & - - s the s i i n no γ γ & 00 \\ s the s i i n no γ γ & cos cos γ γ & 00 \\ 00 & 00 & 11 \end{matrix}] - - - - - - ((99))$

$R R ((β β)) = = [\begin{matrix} c c o o s the s β β & 00 & s the s i i n no β β \\ 00 & 11 & 00 \\ - - sin sin β β & 00 & c c o o s the s β β \end{matrix}] - - - - - - ((1010))$

$R R ((α α)) = = [\begin{matrix} 11 & 00 & 00 \\ 00 & c c o o s the s α α & - - s the s i i n no α α \\ 00 & s the s i i n no α α & cos cos α α \end{matrix}] - - - - - - ((1111))$

Claims

1. A one-stop calibration method for setting parameters of a vehicle-mounted three-dimensional laser scanning system, characterized in that it comprises the following steps:

Step (1), fixing a certain number of artificial markers with reflective properties on the objects in the calibration field, and obtaining the geocentric coordinates of these artificial marker points;

Step (2), select a certain position of the marker point as a site, let the vehicle carrying the 3D laser scanning system stop at the site, use the laser scanning system to scan the object in 3D, and collect a series of marker point clouds;

Step (3), identifying and locating these mark point clouds, and obtaining their coordinates in the laser scanning coordinate system L;

Step (4), the geocentric coordinates of the artificial marks are converted from the geocentric coordinate system (WGS84) to the inertial platform coordinate system (I) through coordinate transformation, and the coordinates in the laser scanning coordinate system (L) are realized through the coordinate transformation model Transform to obtain the placement parameters.

2. The one-stop calibration method for the placement parameters of a vehicle-mounted three-dimensional laser scanning system according to claim 1, wherein the steps of identifying and locating the point cloud of the sign in step (3) include: 1) by The landmark point cloud is subjected to threshold segmentation, 2) clustering the landmark point cloud falling within the receiving threshold, 3) eliminating the noise target, 4) determining the barycentric coordinates of the landmark point in the laser scanning coordinate system.

3. The one-stop calibration method for the placement parameters of a vehicle-mounted three-dimensional laser scanning system according to claim 2, characterized in that: during the threshold segmentation process, the marker points falling within the acceptance threshold are reserved, and the threshold value varies with the marker The distance and angle of the back light reflection of the point vary.

4. A one-stop calibration method for the placement parameters of a vehicle-mounted three-dimensional laser scanning system according to claim 2, characterized in that: the initial center of gravity of the marker points is set, and the marker point cloud is classified one by one to calculate the center of gravity of the marker points. Perform reconstruction to achieve self-growth clustering.