LiDAR point cloud precision enhancement method based on POS trajectory correction

Abstract

The invention relates to a LiDAR point cloud precision enhancement method based on POS trajectory correction, and the method comprises the steps: laying a target control point, and obtaining data; fusing the base station GNSS data, the mobile station GNSS data and the inertial navigation measurement data to obtain an initial POS trajectory; importing actual measurement coordinates of a target control point on the basis of the laser scanning data and the initial POS trajectory, and fusing nearby point clouds to obtain pre-fused point clouds; matching and fitting to obtain a fitting center point; searching the nearest point cloud, and interpolating the time of the fitting center point according to the GPS time information; and correcting the initial POS trajectory according to the actually measured coordinates of the two target control points obtained at the same time and the coordinates of the point cloud target, and performing secondary fusion with the laser scanning data to obtain result data. According to the method, the automatic target matching and center point fitting algorithm is adopted, the error introduction of manual target point selection is reduced, the strict requirement for field data acquisition is reduced, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile three-dimensional laser scanning measurement, and in particular to a LiDAR point cloud accuracy enhancement method based on POS trajectory correction. Background Art

[0002] 3D laser scanning technology is also known as real-scene replication technology. It uses the principle of laser ranging to quickly reconstruct a 3D model of the measured target by recording the 3D coordinates, reflectivity, texture and other information of a large number of dense points on the surface of the measured object at high speed. This technology breaks through the traditional single-point measurement mode and has the advantages of high efficiency, high precision, digitization and automation. It is widely used in the fields of construction, transportation, water conservancy, archaeology, etc.

[0003] For professional surveying fields such as resurveying existing railway lines and expanding and reconstructing highways, high-precision (centimeter-level) point cloud data is usually required. The position accuracy of conventional three-dimensional laser scanning measurement mainly depends on the positioning information of GNSS (Global Navigation Satellite System) and the real-time attitude angle of IMU (Inertial measurement unit) for combined navigation and solution. The accuracy is generally at the decimeter level. In order to meet the high-precision measurement requirements of actual projects, external control points are usually introduced to correct the results of three-dimensional laser scanning to improve the accuracy of the final results. Using control points to correct the accuracy of three-dimensional laser scanning is usually to directly correct the point cloud results, that is, first obtain three-dimensional point cloud data according to conventional data processing, and then calculate the conversion parameters based on two sets of coordinates (point cloud coordinates and measured coordinates) of multiple control points, and finally use the conversion parameters to convert all point clouds. However, this method has a good improvement effect on the point cloud near the control points, but the accuracy of the point cloud between the control points is difficult to guarantee, and it is easy to be distorted or misaligned, which destroys the overall rigid structure and is difficult to meet the needs of high-precision and high-smoothness measurement of traffic engineering.

[0004] Therefore, it is necessary to propose new measures to overcome the above-mentioned defects. Summary of the invention

[0005] The purpose of the present invention is to provide a LiDAR point cloud accuracy enhancement method based on POS trajectory correction to solve the problem that the point cloud accuracy between control points is difficult to ensure in the existing methods.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A LiDAR point cloud accuracy enhancement method based on POS trajectory correction, the method comprising:

[0008] Lay out target control points, obtain base station GNSS data, mobile station GNSS data, inertial navigation measurement data, laser scanning data and measured coordinates of target control points;

[0009] The initial POS trajectory line is obtained by integrating the base station GNSS data, mobile station GNSS data and inertial navigation measurement data;

[0010] Based on the laser scanning data and the initial POS trajectory line, the measured coordinates of the target control point are imported, the fusion section is divided, and the point cloud in the vicinity of the target control point is fused to obtain the pre-fused point cloud;

[0011] Matching and fitting are performed based on the pre-fused point cloud to obtain the fitting center point;

[0012] Search for the point cloud closest to the fitting center point, and interpolate the time of the fitting center point based on its GPS time information;

[0013] The initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time to obtain a new POS trajectory line;

[0014] The new POS trajectory is used to perform secondary fusion on the laser scanning data to obtain the final point cloud data.

[0015] Furthermore, target control points are deployed, including:

[0016] Target control points are arranged in pairs on both sides of the line with a spacing no greater than a specified distance, and the target control points are provided with square or circular target templates.

[0017] Furthermore, base station GNSS data, mobile station GNSS data, inertial navigation measurement data, laser scanning data and measured coordinates of target control points are obtained, including:

[0018] Synchronously set up a GNSS base station at the offline CPI control point or CPII control point to obtain the base station GNSS data;

[0019] Use the GNSS receiver of the mobile 3D laser scanning system to obtain the mobile station GNSS data;

[0020] Use the high-precision inertial navigation measurement equipment of the mobile 3D laser scanning system to obtain inertial navigation measurement data;

[0021] Using a mobile 3D laser scanning system to obtain laser scanning data;

[0022] The actual coordinates of the target control points are obtained by using GNSS rapid static measurement and leveling methods.

[0023] Furthermore, the fusion section is divided into:

[0024] Based on the approximate location of the target control point, a buffer is established at a specified distance or radius to construct a fusion area.

[0025] Furthermore, the point cloud in the vicinity of the target control point is fused to obtain a pre-fused point cloud, including:

[0026] Based on the initial POS trajectory, the coordinate rotation matrix is ​​constructed to fuse the laser scanning data and obtain the pre-fused point cloud.

[0027] Furthermore, matching and fitting are performed based on the pre-fused point cloud to obtain the fitting center point, including:

[0028] In the pre-fused point cloud, the sliding window is used to match the point cloud, and the best matching position is found by calculating the correlation coefficient;

[0029] After accurate registration with the target in the point cloud, the center coordinates of the target template at the target control point are used as the center of the point cloud target. The center point elevation value of the point cloud target is the average of the nearest specified number of point cloud elevations to obtain the fitting center point.

[0030] Furthermore, the initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including:

[0031] According to the angle between the line connecting the measured coordinates of the two target control points obtained at the same time and the line connecting the point cloud target coordinates, the attitude angle of the initial POS trajectory line is corrected, including the heading angle, roll angle, and pitch angle.

[0032] Furthermore, the initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including:

[0033] According to the position deviation between the measured coordinates of the two target control points obtained at the same time and the point cloud target coordinates, the position of the initial POS trajectory line is corrected.

[0034] Furthermore, the initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including:

[0035] The initial POS trajectory line between the measured coordinates of the target control point and the point cloud target coordinates is smoothed.

[0036] Furthermore, the laser scanning data is fused again using the new POS trajectory line, and the result is the WGS84 coordinate system. The point cloud is converted to the engineering independent coordinate system using the WGS84 geodetic height result and the engineering coordinate system normal height result.

[0037] Compared with the prior art, the present invention has the following beneficial effects:

[0038] The present invention provides a LiDAR point cloud accuracy enhancement method based on POS trajectory correction, which solves the problem of poor applicability and accuracy of traditional point cloud correction, adopts target automatic matching and center point fitting algorithms, reduces the error introduced by manually selecting target points, and reduces the stringent requirements for field data acquisition (general operations require slow scanning speed and sufficient point clouds on the target), thereby improving overall operation efficiency. In addition, the present invention adopts a two-way combined navigation algorithm when correcting the POS trajectory line, and smoothes the POS trajectory line through forward and reverse solutions, further improving the overall relative positioning accuracy, and effectively avoiding the distortion and dislocation caused by the destruction of the rigid body of the original object by the control point correction. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0040] Figure 1 It is a flow chart of the method of the present invention.

[0041] Figure 2 It is a schematic diagram of the target control point pattern 1 of the present invention.

[0042] Figure 3 It is a schematic diagram of target control point pattern 2 of the present invention.

[0043] Figure 4 Schematic diagram of the target template and point cloud matching process of the present invention. (a) describes the target matching process, using a template to match the target point cloud by moving horizontally and vertically; (b) describes the result of successful matching, where point P is the center position of the fitted target.

[0044] Figure 5 It is a schematic diagram of the target correction POS error source of the present invention.

[0045] Figure 6 It is a schematic diagram of the two-way combined navigation of the present invention. DETAILED DESCRIPTION

[0046] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. The preferred embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thoroughly understood.

[0047] It should be noted that the terms "comprises", "comprising", "includes", "comprises" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0048] In addition, steps "S1", "S2" and the like involved in the embodiments are only for the convenience of describing the technical solutions and understanding the specific steps and contents of the embodiments, and should not be understood as limitations on the order of the steps. Any changes to the order of the steps should be within the scope of protection of the present invention.

[0049] Aiming at the high-precision measurement requirements of existing transportation projects, the present invention proposes a LiDAR point cloud accuracy enhancement method based on POS trajectory correction. Before generating point cloud data results, the initial POS trajectory is first solved and processed using the deployed target control points to correct its position and attitude errors to obtain a high-precision corrected trajectory, which is then fused with the scanning data to obtain the final point cloud result. This method can improve the point cloud accuracy without destroying the overall rigidity, effectively ensure the absolute and relative accuracy of the point cloud, and meet the high-precision measurement requirements of transportation projects.

[0050] Specifically, Figure 1 The LiDAR point cloud accuracy enhancement method based on POS trajectory correction provided by the present invention specifically includes the following steps:

[0051] S1: Measurement data acquisition, including acquisition of base station GNSS data, mobile station GNSS data, inertial navigation measurement data, laser scanning data and measured coordinates of target control points.

[0052] In response to the measurement needs of railways or highways, this method uses a mobile 3D laser scanning system to acquire data. Before the scanning operation, target control points are set up on both sides of the line in advance, and GNSS base stations are set up on the target control points for synchronous observation. After the scanning operation is completed, the original data is copied and sorted. The mobile 3D laser scanning measurement system used can use a vehicle-mounted or airborne platform equipped with a 3D laser scanner to scan the traffic engineering area according to actual conditions. The laser scanner has a point frequency of not less than 1MHz / s and is equipped with a high-precision inertial navigation measurement device. The vehicle-mounted platform can also add an odometer device.

[0053] The specific process of setting up target control points is: set up target control points in pairs on both sides of the line with a spacing of no more than 500 meters, and the target control points have square or circular target templates. Figure 2 and Figure 3 As shown, it can be designed into a rectangle or circle (side length or diameter not less than 50cm), made of foam board (thickness not less than 5mm), spray-painted in black and white, and finally coated with a layer of frosted film to increase the reflectivity of the laser.

[0054] The data include:

[0055] 1. Synchronously set up a GNSS base station at the offline CPI control point or CPII control point. The base station coverage distance is no more than 10km. Usually, multiple base stations are deployed along the line to obtain the base station GNSS data;

[0056] 2. Use the GNSS receiver of the mobile 3D laser scanning system to obtain the GNSS data of the mobile station;

[0057] 3. Use the high-precision inertial navigation measurement equipment of the mobile 3D laser scanning system to obtain inertial navigation measurement data;

[0058] 4. Use a mobile 3D laser scanning system to obtain laser scanning data;

[0059] 5. Use GNSS rapid static measurement and leveling methods to obtain the measured coordinates of the target control points.

[0060] In this step, the base station control network can give priority to the use of the existing CPI and CPII control networks of the railway. When the point density and accuracy of the control network do not meet the survey requirements, it is necessary to build a new base station control network along both sides of the line.

[0061] S2: Use the combined navigation solution software to integrate the base station GNSS data, mobile station GNSS data, and inertial navigation measurement data to obtain the initial POS trajectory line and the initial scanning trajectory line file, check the quality factor, position accuracy and other indicators, and check the solution accuracy of the trajectory line.

[0062] This step mainly uses integrated navigation solution software (such as Inertial Explorer) to perform differential analysis on the above data. The solution strategy selects Tightly Coupled. The selection of multiple base stations and elimination of constellations with poor quality can improve the solution accuracy.

[0063] S3: Based on the laser scanning data and the initial POS trajectory line, the measured coordinates of the target control point are imported, and the fusion section is divided. With the target position as the center, the point cloud in the vicinity of the target control point is fused. The radius or distance can be specified to obtain the pre-fused point cloud. This step only fuses the point cloud in the vicinity of the target to improve the efficiency of subsequent target matching.

[0064] S301: Dividing the fusion section, including:

[0065] Constructs the blend region by creating a buffer zone at a specified distance or radius based on the approximate location of the target control point.

[0066] S302: Fusing the point cloud in the vicinity of the target control point to obtain a pre-fused point cloud, including:

[0067] Based on the initial POS trajectory, the laser scanning data is fused and solved by constructing a coordinate rotation matrix to obtain a pre-fused point cloud. The principle of point cloud fusion is:

[0068] P XYZ =R P *(R d *P I +T d )+T P

[0069] in:

[0070] P XYZ The absolute coordinates of the point cloud after fusion and solution;

[0071] P I The original coordinates of the original laser point cloud scan at the time of recording;

[0072] T P (X P ,Y P ,Z P ) is the POS trajectory position at that moment obtained by time linear interpolation;

[0073] T d (X d ,Y d ,Z d ) is the offset of the laser original coordinate system relative to the inertial navigation coordinate system recorded in the system calibration file;

[0074] R P and R d is the rotation matrix.

[0075] If all point clouds are fused, the data volume will be relatively large, which will bring a lot of workload for target matching in step S4. Therefore, this method superimposes the measured coordinates of the target, divides the POS trajectory line into multiple segments according to the position of the target, and only fuses the point cloud of the POS segment, which can not only fuse the target point cloud but also reduce the data volume.

[0076] S4: Match and fit based on the pre-fused point cloud to obtain the fitting center point.

[0077] Automatic target matching and center point fitting are to solve the problem of correcting the target point selection. Since the points emitted by the three-dimensional laser are random, they may just fall on the center of the physical target. However, in most cases, there is no scanned point cloud at the center of the target. Usually, points near the center are selected as substitutes when correcting the point selection, which will introduce artificial errors. Step S4 is to define the same size as the physical target as a template, match it with the point cloud in the pre-fused point cloud by sliding a window, perform matching calculations from the point cloud according to a certain step length, and find the best matching position by calculating the correlation coefficient. At this time, the theoretical center position of the target template is considered to be the center of the target in the point cloud. This method continues to select the average of the Z coordinates of the five point clouds closest to the center point as the elevation of the theoretical center point, so that the three-dimensional coordinates of the center point of the fitted target can be determined.

[0078] This step specifically includes:

[0079] S401: Matching the pre-fused point cloud with the point cloud by means of a sliding window, such as Figure 2 and Figure 3 , find the best matching position by calculating the correlation coefficient;

[0080] S402: After accurate registration with the target in the point cloud, the center coordinates of the target template at the target control point are used as the center of the point cloud target, and the elevation value of the center point of the point cloud target is the average of the elevations of the nearest specified number of point clouds to obtain the fitting center point. The three-dimensional coordinates of the fitting center point in the point cloud framework are recorded.

[0081] S5: Search for the point cloud closest to the fitting center point, and interpolate the time of the fitting center point according to its GPS time information.

[0082] This step is to assign GPS time to the target center point in order to establish a synchronous relationship between the fitting center point and the POS trajectory line. Since the target center point is fitted and not the point actually scanned by the three-dimensional laser, there is no GPS time information and it cannot correspond to the POS trajectory line. It is difficult to establish a corrective relationship later. It is necessary to perform multi-dimensional interpolation based on the GPS time of the point cloud near the fitting center point to assign time information to the target center point. Usually, the point frequency of the laser scanner is very high (1 million points / second), and the time difference between adjacent point clouds is about 0.01ms. The time of the fitting point can be interpolated based on its GPS time information by searching for the point cloud closest to the fitting center point, thereby assigning timestamp information to the target center point.

[0083] S6: Correct the initial POS trajectory line according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time to obtain a new POS trajectory line.

[0084] After all the target points in each measurement section are matched and fitted, each target point has two sets of coordinates (measured coordinates, coordinates in the point cloud) and GPS time information. According to the two sets of coordinates of each target point and the recorded time information, a correction matrix is ​​constructed to correct the angle and coordinates of the POS trajectory line at the same time. After the correction is completed, the POS trajectory line is smoothed using a two-way combined navigation algorithm to obtain the final corrected high-precision POS trajectory line. Specifically, the correction includes the following three aspects:

[0085] S601: POS posture correction. After the above steps, multiple sets of target coordinates are obtained. Assuming that each pair of targets is scanned at the same time, the actual target center connection line ( Figure 5 The black line in the figure) and the point cloud fitting target center line ( Figure 5 The red line in the figure is theoretically overlapped. Due to the attitude error, there will be an angle between the two lines. By decomposing the above angles (into heading angle, roll angle, and pitch angle), the POS attitude at that moment can be corrected. Specifically, according to the angle between the measured coordinates of the two target control points obtained at the same time and the point cloud target coordinates, the attitude angle of the initial POS trajectory line is corrected, including the heading angle, roll angle, and pitch angle.

[0086] S602: POS position correction. There may be a deviation between the actual position of the target and the point cloud fitting position ( Figure 5 The deviation can be decomposed into three direction values: east, north, and altitude. The POS position is corrected using the deviation values. Specifically, the position of the initial POS trajectory line is corrected based on the position deviation between the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time.

[0087] S603: POS smoothing. In the actual operation process, multiple pairs of targets will be arranged at a certain interval. The deviation of each pair of targets can be used to correct the POS at that moment. If the area in the middle of the target is not processed, the final point cloud will be distorted and stepped, and the relative accuracy will deteriorate. Therefore, it is necessary to smooth the POS in the target interval. The two-way combined navigation algorithm can be used to fix the corrected POS of the targets at both ends, and smooth the POS according to the forward and reverse calculations ( Figure 6 ), correct the error of the interval segment and ensure smoothness. Specifically, the initial POS trajectory line between the measured coordinates of the target control point and the point cloud target coordinates is smoothed.

[0088] S7: Use the new POS trajectory line to perform secondary fusion of the laser scanning data, and obtain the final point cloud data through processes such as coordinate system conversion, format conversion, and accuracy verification.

[0089] Using the new POS trajectory line, the laser scanning data is fused again, and the result is the WGS84 coordinate system. Using the WGS84 geodetic height result and the normal height result of the engineering coordinate system, the point cloud is converted to the engineering independent coordinate system.

[0090] Example:

[0091] In a resurvey project of an existing railway, airborne LiDAR measurement was used to acquire data. Target plates (50×50cm rectangle) were arranged in pairs at 300-meter intervals. After data preprocessing, the point cloud accuracy enhancement method of the present invention was used to efficiently and accurately complete the POS track line correction work. The errors in the final point cloud plane and elevation were both within 1 cm. The three-dimensional coordinates of the track centerline were extracted using the point cloud. After professional fitting by the design department, the plane met the requirements of the line design specifications, and the elevation also met the train smoothness requirements. Since target correction is no longer restricted by point cloud density, the flight speed of this project was increased from 8m / s to 12m / s, greatly improving the work efficiency.

[0092] This patented method has a simple technical route, rigorous theory, and strong operability. It can improve the accuracy of the point cloud without destroying the overall rigidity, effectively ensure the absolute and relative accuracy of the point cloud, and meet the high-precision measurement needs of transportation engineering. It is suitable for point cloud accuracy enhancement of mobile scanning measurement on any platform, and can meet the needs of high-precision measurement in different industries and fields.

[0093] The above specific examples are used to illustrate the present invention, which is only used to help understand the present invention and is not intended to limit the present invention. For those skilled in the art, according to the concept of the present invention, some simple deductions, modifications or substitutions can be made.

Claims

1. A LiDAR point cloud accuracy enhancement method based on POS trajectory correction, characterized by: The method comprises: Lay out target control points, obtain base station GNSS data, mobile station GNSS data, inertial navigation measurement data, laser scanning data and measured coordinates of target control points; The initial POS trajectory line is obtained by integrating the base station GNSS data, mobile station GNSS data and inertial navigation measurement data; Based on the laser scanning data and the initial POS trajectory line, the measured coordinates of the target control point are imported, the fusion section is divided, and the point cloud in the vicinity of the target control point is fused to obtain the pre-fused point cloud; Matching and fitting are performed based on the pre-fused point cloud to obtain the fitting center point; Search for the point cloud closest to the fitting center point, and interpolate the time of the fitting center point based on its GPS time information; The initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time to obtain a new POS trajectory line; The new POS trajectory is used to perform secondary fusion on the laser scanning data to obtain the final point cloud data.

2. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Laying out target control points, including: Target control points are arranged in pairs on both sides of the line with a spacing no greater than a specified distance, and the target control points are provided with square or circular target templates.

3. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Acquire base station GNSS data, mobile station GNSS data, inertial navigation measurement data, laser scanning data and measured coordinates of target control points, including: Synchronously set up a GNSS base station at the offline CPI control point or CPII control point to obtain the base station GNSS data; Use the GNSS receiver of the mobile 3D laser scanning system to obtain the mobile station GNSS data; Use the high-precision inertial navigation measurement equipment of the mobile 3D laser scanning system to obtain inertial navigation measurement data; Using a mobile 3D laser scanning system to obtain laser scanning data; The actual coordinates of the target control points are obtained by using GNSS rapid static measurement and leveling methods.

4. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Divide the fusion section, including: Based on the approximate location of the target control point, a buffer is established at a specified distance or radius to construct a fusion area.

5. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Fuse the point clouds in the vicinity of the target control point to obtain a pre-fused point cloud, including: Based on the initial POS trajectory, the coordinate rotation matrix is ​​constructed to fuse the laser scanning data and obtain the pre-fused point cloud.

6. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Matching and fitting are performed based on the pre-fused point cloud to obtain the fitting center point, including: In the pre-fused point cloud, the sliding window is used to match the point cloud, and the best matching position is found by calculating the correlation coefficient; After accurate registration with the target in the point cloud, the center coordinates of the target template at the target control point are used as the center of the point cloud target. The center point elevation value of the point cloud target is the average of the nearest specified number of point cloud elevations to obtain the fitting center point.

7. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: The initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including: According to the angle between the line connecting the measured coordinates of the two target control points obtained at the same time and the line connecting the point cloud target coordinates, the attitude angle of the initial POS trajectory line is corrected, including the heading angle, roll angle, and pitch angle.

8. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: The initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including: According to the position deviation between the measured coordinates of the two target control points obtained at the same time and the point cloud target coordinates, the position of the initial POS trajectory line is corrected.

9. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: The initial POS trajectory line is corrected according to the measured coordinates of the two target control points and the point cloud target coordinates obtained at the same time, including: The initial POS trajectory line between the measured coordinates of the target control point and the point cloud target coordinates is smoothed.

10. The LiDAR point cloud accuracy enhancement method based on POS trajectory correction according to claim 1, characterized in that: Using the new POS trajectory line, the laser scanning data is fused again, and the result is the WGS84 coordinate system. Using the WGS84 geodetic height result and the normal height result of the engineering coordinate system, the point cloud is converted to the engineering independent coordinate system.

Images