Digital elevation model construction method based on unmanned aerial vehicle laser radar

Abstract

The invention relates to the technical field of unmanned aerial vehicle laser point cloud, and discloses a digital elevation model construction method based on an unmanned aerial vehicle laser radar, and the method comprises the steps: carrying out the field flight of a target region based on an unmanned aerial vehicle carrying a laser radar load, and obtaining the point cloud data of the laser radar; and based on the laser radar point cloud data, carrying out point cloud data calculation to generate a digital elevation model. The technical means that the unmanned aerial vehicle carries a laser radar load is adopted, field flight is carried out on an east area salt pan sedimentation basin area to obtain laser radar point cloud data, the instrument integrates a high-precision laser scanner and an inertial navigation system, the overall structure is small, the weight is light, and the instrument is suitable for data collection on an unmanned aerial vehicle platform; high-precision three-dimensional data of the earth surface can be rapidly collected in real time, and point cloud data calculation is carried out through digital green soil LiDAR360 software to generate a digital elevation model.

Description

Technical Field

[0001] The present application relates to the field of unmanned aerial vehicle laser point cloud technology, and in particular to a method for constructing a digital elevation model based on unmanned aerial vehicle laser radar. Background Art

[0002] Digital Elevation Model (DEM) is a digital simulation of ground terrain (i.e., a digital expression of terrain surface morphology) achieved through limited terrain elevation data. It is a physical ground model that represents ground elevation in the form of a set of ordered numerical arrays. It is a branch of Digital Terrain Model (DTM), and various other terrain characteristic values ​​can be derived from it.

[0003] It is generally believed that DTM describes the spatial distribution of linear and nonlinear combinations of various geomorphic factors including elevation, such as slope, aspect, slope change rate and other factors. DEM is a zero-order simple single-item digital geomorphic model. Other geomorphic characteristics such as slope, aspect and slope change rate can be derived based on DEM.

[0004] There are many ways to build a DEM. From the data source and collection method, there are: (1) direct measurement from the ground, the instruments involved include horizontal guide rails, probes, probe holders and relative elevation measurement plates, etc., and high-end instruments such as GPS, total stations, and field measurements can also be used; (2) based on aerial or aerospace images, through photogrammetry, such as stereo coordinate instrument observation and aerial triangulation encryption, analytical mapping, digital photogrammetry, etc.; (3) from existing topographic maps, such as grid reading method, digitizer hand-held tracking and scanner semi-automatic collection and then interpolation to generate DEM and other methods. There are many DEM interpolation methods, mainly including global interpolation, block interpolation and point-by-point interpolation. The fitting model of global interpolation is established by the observation values ​​of all sampling points in the study area. Block interpolation is to divide the reference space into several blocks of the same size and use different functions for each block. Point-by-point interpolation is centered on the point to be interpolated and defines a local function to fit the surrounding data points. The range of data points changes with the change of the position to be interpolated, so it is also called moving fitting method. There are two algorithms: regular network structure and irregular triangulated network (Triangular Irregular Network, referred to as TIN).

[0005] In the prior art, CN118279509A discloses a method and device for obtaining a high-precision digital elevation model for terrain research, the method comprising: obtaining different sample plots in a terrain study area, and corresponding radar digital elevation models and unmanned aerial vehicle digital elevation models in the current period, wherein for each of the sample plots, the sample plot is a sub-area that characterizes the terrain and geomorphic features of the terrain study area; constructing a regression relationship model according to the radar digital elevation models and unmanned aerial vehicle digital elevation models corresponding to each of the sample plots, wherein the regression relationship model characterizes the mapping relationship between each radar digital elevation model and each unmanned aerial vehicle digital elevation model; based on the regression relationship model, correcting the radar digital elevation model corresponding to the terrain study area; the method has the effect of improving the precision of the digital elevation model and the accuracy of the data generated by the digital elevation model.

[0006] After searching, it is found that there is no technical solution that combines field data collection and indoor data processing to improve the accuracy of the constructed digital elevation model. Therefore, this application combines field data collection and indoor data processing to improve the accuracy of the digital elevation model. Summary of the invention

[0007] The purpose of this application is to provide a method for constructing a digital elevation model based on UAV lidar, by combining field data collection and indoor data processing to improve the accuracy of the digital elevation model.

[0008] In order to achieve the above purpose, the technical solutions adopted are as follows:

[0009] A method for constructing a digital elevation model based on unmanned aerial vehicle laser radar, the method comprising:

[0010] Based on the UAV equipped with LiDAR payload, fly over the target area to obtain LiDAR point cloud data;

[0011] Based on the laser radar point cloud data, point cloud data is solved to generate a digital elevation model.

[0012] Furthermore, based on the UAV equipped with the LiDAR payload, a field flight is conducted over the target area to obtain LiDAR point cloud data, including:

[0013] The goal is to obtain the LiDAR point cloud data of the target area before and after cleaning, and design the flight range and route of the UAV;

[0014] Determine the take-off and landing site of the drone according to the requirements of the drone equipment operation manual;

[0015] The UAV is controlled to perform field flight according to a set flight control strategy to obtain laser radar point cloud data.

[0016] Furthermore, the flight control strategy is set, including:

[0017] Flight control method;

[0018] Aviation and route curvature control methods.

[0019] Furthermore, the flight control method includes:

[0020] During the entire flight, the aircraft's turning slope is controlled within 15°;

[0021] During the flight along the survey line, the following requirements must be met:

[0022] Different routes use alternating left and right turns;

[0023] The deviation from the route does not exceed 25m, and the altitude does not deviate by 25m;

[0024] The yaw angle is not greater than 6°;

[0025] The tilt angle of the image is not greater than 2°;

[0026] The curvature of the route is not greater than 3%;

[0027] The aircraft speed is maintained at the designed speed for aerial photography.

[0028] Further, the aviation and route curvature control methods include:

[0029] The flight altitude is monitored in real time through the LiDAR measurement system, and corrections are made when the actual flight altitude value deviates from the designed flight altitude.

[0030] Furthermore, based on the laser radar point cloud data, point cloud data is solved to generate a digital elevation model, including:

[0031] Import the laser radar point cloud data into the Qianxunyi laser radar system-point cloud preprocessing module for format conversion, and use the converted IMR data and GNSS data to upload to the Qianxun Cloud Control Console to calculate the trajectory data and radar data;

[0032] Input the solved trajectory data and radar data into the Qianxunyi LiDAR System-Point Cloud Preprocessing Module, input the engineering path, coordinate system, and set the point cloud generation parameters and filtering parameters in turn to solve the point cloud, and automatically generate high-precision point cloud data;

[0033] Based on the high-precision point cloud data, the required data is selected, the resolution and DEM construction are set, a triangulated network is constructed for key points, and after interpolation processing, the grid is transformed into raster data to obtain a digital elevation model.

[0034] Furthermore, the solved trajectory data and radar data are input into the Qianxunyi LiDAR System-Point Cloud Preprocessing Module, and the engineering path, coordinate system, and point cloud generation parameters and filtering parameters are input in turn for point cloud solution, automatically generating high-precision point cloud data, including:

[0035] Layered inspection;

[0036] Point cloud denoising;

[0037] Point cloud post-processing.

[0038] Further, layered examinations include:

[0039] The profile and quality inspection tools of the Qianxunyi LiDAR System-Point Cloud Post-Processing Module are used to detect the stratification between routes. If stratification exists, flight strip adjustment is used to reduce the stratification phenomenon.

[0040] Furthermore, point cloud denoising includes removing invalid point clouds and unreasonable point clouds.

[0041] Further, point cloud post-processing includes:

[0042] Point cloud filtering: Construct a triangulated network through the surface point cloud and remove the foot points of the ground objects in the point cloud; classify the ground point cloud based on the iterative morphological filtering method, first divide the survey area into blocks, and the minimum size of the survey area block is larger than the maximum building size in the area, then select the lowest point and the ground points within the set height difference range from the point cloud as ground seed points, establish the initial sparse triangulated network, calculate the angle and distance between the remaining points in the point cloud and the triangulated network, and compare them with the set iteration angle and iteration distance. If it is less than the set iteration value, it is accepted as a ground point, otherwise it is rejected as a ground point, and the above iterations are repeated until no new ground points are added to the triangulated network, thus realizing the process of triangulating network encryption;

[0043] Point cloud editing and classification: remove gross points that are not filtered out after automatic filtering, identify points that are not correctly classified by automatic classification, and add elevation feature points in areas with terrain distortion.

[0044] The beneficial effects of this application are:

[0045] This application uses the technical means of using a drone equipped with a lidar payload to conduct on-site flights over the salt field sedimentation pond area in the eastern district to obtain lidar point cloud data. The instrument integrates a high-precision laser scanner and an inertial navigation system. It has a small overall structure and light weight, and is suitable for data collection on a drone platform. It can quickly and in real time collect high-precision three-dimensional surface data. The internal industry uses the Digital Green Earth LiDAR360 software to solve the point cloud data and generate a digital elevation model (DEM). BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1An overall flow chart of a method for constructing a digital elevation model based on a UAV laser radar provided in an embodiment of the present application;

[0047] Figure 2 A schematic diagram of the measurement area provided in the embodiment of the present application;

[0048] Figure 3 QHCORS distribution diagram provided for the embodiments of the present application;

[0049] Figure 4 A schematic diagram of the Matrice 300 RTK UAV platform provided in the embodiments of the present application;

[0050] Figure 5 The laser radar module parameters provided in the embodiment of the present application;

[0051] Figure 6 A flowchart of the laser radar aerial survey operation provided in the embodiment of the present application;

[0052] Figure 7 A schematic diagram of route planning before cleaning the East Salt Field Settling Tank provided in an embodiment of the present application;

[0053] Figure 8 A schematic diagram of route planning after cleaning of the East Salt Field Settling Tank provided in the embodiment of the present application;

[0054] Fig. 9 A schematic diagram of the take-off and landing points of a drone provided in an embodiment of the present application;

[0055] Fig.10 A schematic diagram of a drone field flight provided in an embodiment of the present application;

[0056] Fig.11 A schematic diagram of the point cloud trajectory before cleaning the East Salt Field Settling Tank provided in an embodiment of the present application;

[0057] Fig.12 A schematic diagram of the point cloud trajectory after cleaning the East District Salt Field Settling Tank provided in the embodiment of the present application;

[0058] Fig.13 A point cloud solution flow chart provided in an embodiment of the present application;

[0059] Fig.14 Point cloud layered inspection diagram before cleaning of the East District Salt Field Settling Tank provided in the embodiment of the present application;

[0060] Fig.15 Point cloud layered inspection diagram after cleaning of the East District Salt Field Settling Tank provided in the embodiment of the present application;

[0061] Fig.16 A schematic diagram of point cloud denoising before cleaning the East District Salt Field Settling Tank provided in an embodiment of the present application;

[0062] Fig.17 A schematic diagram of point cloud denoising after cleaning the East District Salt Field Settling Tank provided in an embodiment of the present application;

[0063] Fig.18 A DEM schematic diagram of the East District Salt Field Settling Tank before cleaning provided in the embodiment of the present application;

[0064] Fig.19 A DEM diagram of the East District Salt Field Settling Tank after cleaning provided in the embodiment of the present application;

[0065] Fig. 20 A schematic diagram of point cloud density calculation before cleaning of the East Salt Field Settling Tank provided in an embodiment of the present application;

[0066] Fig.21 A schematic diagram of point cloud density calculation after cleaning of a salt pan sedimentation tank provided in an embodiment of the present application;

[0067] Fig. 22 The elevation accuracy inspection diagram of the East District Salt Field Settling Tank before cleaning provided in the embodiment of the present application;

[0068] Fig.23 This is an elevation accuracy inspection diagram of the East District Salt Field Settling Tank after cleaning provided in an embodiment of the present application. DETAILED DESCRIPTION

[0069] The following describes the embodiments of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. The present application can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0070] The specific implementation of the present application is further described in detail below in conjunction with the drawings and examples.

[0071] See also Figure 1 , is an overall flow chart of a method for constructing a digital elevation model based on a UAV laser radar provided in an embodiment of the present application. The embodiment of the present application provides a method for constructing a digital elevation model based on a UAV laser radar, and the method comprises the following steps:

[0072] S1: Based on the UAV equipped with LiDAR payload, the target area is flown on the spot to obtain LiDAR point cloud data;

[0073] S2: Based on the LiDAR point cloud data, the point cloud data is solved to generate a digital elevation model.

[0074] In this embodiment, a specific case is provided. The target area is located in the Golmud City area of ​​Haixi Prefecture, Qinghai Province, in the south-central part of the Qaidam Basin, and is under the jurisdiction of Golmud City, Qinghai Province. Geographic coordinates: 95°16′27″ east longitude, 36°47′50″ north latitude.

[0075] From Golmud, drive 60km along the Chag Expressway to reach Chaerhan, and drive 7.2km northeast along the shortcut to reach the survey area. The traffic is generally convenient. Figure 2 .

[0076] The QHCORS signal covers the entire survey area, using the QHCORS station data (Gaersu Station) of the Qinghai Province Satellite Positioning Continuous Operation Service System. The coordinate system is the CGCS2000 National Geodetic Coordinate System and the 1985 National Height Datum. For details of the QHCORS distribution map, see Figure 3 .

[0077] There are geographical control points buried in years and months around the survey area, see Table 1 for details. After field investigation, YYTG03, YYTG05 and YYTG06 are well preserved and can be used as reference points for verification after inspection. The inspection records are shown in Table 2 for details.

[0078] Table 1 Results of topographic control points

[0079]

[0080] Table 2 Field inspection record

[0081]

[0082] The basis for this specific case study is as follows:

[0083] 1. "Urban Surveying Specification" (CJJ / T8-2011);

[0084] 2. “Quality Evaluation Index and Calculation Method of Airborne LiDAR Point Cloud Data” (GB / T36100-2018);

[0085] 3. Technical Specifications for Airborne LiDAR Data Acquisition (CH / T8024-2011);

[0086] 4. Technical Specifications for Airborne LiDAR Data Processing (CH / T8023-2011);

[0087] 5. "Technical Specifications for Real-Time Kinematic Measurement (RTK) of Global Positioning System" CH / T2009-2010;

[0088] 6. Basic geographic information digital achievements: "1:500, 1:1000, 1:2000 digital elevation model" (CH / T9008.2-2010);

[0089] 7. "Quality Inspection and Acceptance of Surveying and Mapping Results" GB / T24356-2009;

[0090] 8. "Regulations on the Preparation of Surveying and Mapping Technical Summary" CH / T 1001-2005;

[0091] 9. "Quality Inspection and Acceptance of Digital Surveying and Mapping Results" GB / T 18316-2008;

[0092] 10. Technical Requirements for UAV Aerial Photography Systems, CH / Z 3002-2010;

[0093] 11. Basic Requirements for Safe Operation of Unmanned Aerial Photography, CH / Z 3001-2010.

[0094] The technical indicators of this specific case are as follows:

[0095] (1) Coordinate system: National 2000 coordinate system;

[0096] (2) Height datum: 1985 National Height Datum;

[0097] (3) Projection mode: Gauss-Krüger 3° zone, zone number 32;

[0098] (4) Central meridian: 96°.

[0099] The configuration of technical personnel and equipment for this specific case is as follows:

[0100] Technical personnel configuration: In order to provide results data in a timely manner, this project has invested 1 working group, with a total of 5 technical management personnel, surveyors, and service personnel participating in this work, including 1 senior surveying and mapping engineer, 2 surveying and mapping engineers, 1 assistant surveying and mapping engineer, and 1 driver.

[0101] The configuration of instruments and equipment is shown in Table 3.

[0102] Table 3 Main hardware configuration table

[0103] Serial number Device Name quantity model Origin use 1 Rotary Wing UAV 1 Matrice 300RTK China Field flight 2 LiDAR 1 QX-200XT China Field flight 3 GPS Receiver 1 STORAGE China Data collection 4 desktop computer 1 unit Assembly Machine China Internal data processing 5 Laptop 1 unit Millet China Field flight 6 External Hard Drive 1 4T China Data backup 7 digital camera 1 unit 8 megapixels China Take work photos 8 Transportation 1 car Off-road vehicles China 9 Track Master 1 DJI China Field flight 10 Digital Green Earth 1 LiDAR360 China Internal data processing 11 Thousand Patrol Wings 1 LiDAR System China Internal data processing 12 ArcGIS Software 1 ArcGIS 10.5 USA Earthwork calculation

[0104] The above instruments and equipment have been inspected, tested and verified in accordance with the regulations and are all within the valid time. UAV and sensor equipment: The Matrice 300 RTK quad-rotor operating platform and the Qianxun QX-200XT laser radar module of Shenzhen DJI Innovations Technology Co., Ltd. are used. Figure 4 Its parameters are as follows: Figure 5 shown.

[0105] This specific case uses the Jingwei M300 RTK drone equipped with a QX-200XT laser radar payload to conduct field flights in the East District Salt Field Settlement Basin area to obtain laser radar point cloud data. The instrument integrates a high-precision laser scanner and an inertial navigation system. The overall structure is small and light, and it is suitable for data collection on a drone platform. It can quickly collect high-precision three-dimensional surface data in real time. The internal industry uses the Digital Green Earth LiDAR360 software to solve the point cloud data and generate a digital elevation model (DEM). Laser radar aerial survey operations such as Figure 6 As shown, the following steps are included:

[0106] Step 1: Flight design.

[0107] In order to obtain image data of the East District Salt Field Sedimentation Basin before and after cleaning, the flight range and route were designed in accordance with the field aerial photogrammetry specifications and volume measurement requirements. Figure 7 and Figure 8 shown.

[0108] Step 2: Field data collection.

[0109] According to the requirements of the UAV equipment operation manual, find a place suitable for UAV take-off and landing, that is, try to find a place with flat terrain, wide vision, and unobstructed take-off and landing points. The take-off and landing point selected for the study is located in the open space on the northeast side of the East District Salt Field Sedimentation Pool area. The specific location is as follows Fig. 9 shown.

[0110] After the take-off and landing site is determined, the relevant equipment is debugged, the flight speed is set to 10 meters per second, and the relative altitude is 80 meters. Fig.10 shown.

[0111] Step 3: Determine quality assurance measures.

[0112] LiDAR production quality management runs through the entire process from task acceptance, organizational preparation, technical design, production operations, process inspection to product delivery and use.

[0113] This embodiment implements a two-level inspection, namely process inspection and final inspection; the process inspection is further divided into self-inspection and mutual inspection; the inspection personnel are composed of the project technical person in charge, the quality inspector of the Geoscience Center and the project person in charge, and the final inspection is reviewed by experts organized by the institute-level technical department.

[0114] Step 4, flight quality assurance.

[0115] (1) Flight control measures

[0116] During the entire flight, the aircraft's turning slope was controlled within 15° (standard turning), avoiding the loss of GPS satellite signal lock. During the flight along the survey line, the following requirements were met:

[0117] Different routes use alternating left and right turns;

[0118] The deviation from the route does not exceed 25m, and the altitude does not deviate by 25m;

[0119] The yaw angle is not greater than 6°;

[0120] The tilt angle of the image is not greater than 2°;

[0121] The curvature of the route is not greater than 3%;

[0122] The aircraft speed is maintained at the designed speed for aerial photography.

[0123] (2) Pre-flight ground testing

[0124] In accordance with strict instrument operating specifications, ground power-on tests and related preparations are carefully performed before each takeoff to ensure that the entire aerial photography system is in normal condition.

[0125] (3) Flight altitude and route curvature control

[0126] Aerial photography operators monitor the flight altitude in real time through the lidar measurement system. When the actual flight altitude value deviates from the designed flight altitude, they communicate with the captain in a timely manner to make corrections, ensuring that the flight altitude meets the design requirements.

[0127] (4) Heading and lateral overlap control

[0128] Since the LiDAR measurement system integrates an advanced aerial photography control system to achieve GPS fixed-point exposure, it ensures that the heading and lateral overlap of the aerial photographs are effectively guaranteed after the aircraft flies along the set route.

[0129] Step 5, internal data processing, includes the following steps:

[0130] Step 51: point cloud data preprocessing.

[0131] Copy the original point cloud, IMU data, GNSS data and other necessary data for internal work obtained in the field to the computer and confirm the integrity of the data. The first step of internal processing is trajectory solution, and its solution accuracy is directly related to the accuracy of the final result.

[0132] First, import the IMU data into the Qianxunyi LiDAR system-point cloud preprocessing module for format conversion. Use the converted IMR data and GNSS data to upload to the Qianxun Cloud Console to calculate the trajectory data, such as Fig.11 and Fig.12As shown in the figure, after calculation, the trajectory accuracy is less than 0.02 meters, the position calculation meets the specification requirements, the attitude angle range is within 0.06 degrees, the heading angle range is within 0.03 degrees, and the attitude calculation is qualified. It can be used as effective data for point cloud generation.

[0133] Step 52, point cloud solution.

[0134] Input the solved trajectory data and radar data into the Qianxunyi LiDAR System-Point Cloud Preprocessing Module, and input the project path, coordinate system, and set the necessary information such as point cloud generation parameters and filter parameters according to the prompts. After confirming that the input information is correct, click Point Cloud Solution, and the software will automatically generate high-precision point cloud data, such as Fig.13 As shown, the solution process is as follows:

[0135] Step 521, layered inspection.

[0136] In order to verify whether there is stratification in the solved point cloud data, the profile and quality inspection tools of the Qianxunyi LiDAR System-Point Cloud Post-Processing Module can be used to quickly detect the stratification between routes (the darker the color, the more obvious the stratification). If there is a more obvious stratification, you can use flight adjustment to weaken the stratification phenomenon and improve data quality. The point cloud data solved this time has no stratification. In accordance with the requirements of the standard, the quality of the radar acquired data is good and can be used for subsequent processing and production. The quality inspection diagram is shown below. See details. Fig.14 and Fig.15 .

[0137] Step 522, point cloud denoising.

[0138] Remove invalid and unreasonable point clouds. Fig.16 and Fig.17 .

[0139] Step 53, point cloud post-processing.

[0140] Point cloud post-processing uses the Qianxunyi LiDAR system-point cloud post-processing module.

[0141] Step 531: point cloud filtering.

[0142] The triangulated network is constructed through the surface point cloud, and the foot points of the objects in the point cloud are removed. This process is called point cloud filtering. The ground point cloud is classified using iterative morphological filtering. This method first divides the survey area into blocks. The minimum size of the survey area block is larger than the size of the largest building in the area. Then, the lowest point and ground points within a certain height difference range are selected from the point cloud as ground seed points to establish the initial sparse triangulated network. The angle, distance and other information of the remaining points in the point cloud to the triangulated network are calculated and compared with the set iteration angle and iteration distance. If it is less than the set iteration value, it is accepted as a ground point. Otherwise, it is rejected as a ground point. The above iterations are repeated until no new ground points are added to the triangulated network, thus realizing the process of triangulating network encryption.

[0143] Step 532, point cloud editing and classification.

[0144] After filtering, most buildings have been separated from the ground point cloud, but some buildings are still mixed in the ground point cloud. In addition, due to building occlusion, the densely built area and the ground point cloud are too sparse, resulting in terrain distortion, which requires manual editing and classification. The main purpose of manual editing is to remove a small number of gross points that are not filtered out after automatic filtering, identify points that are not correctly classified by automatic classification, and appropriately increase the elevation feature points in the terrain distortion area, so as to achieve the purpose of controlling the accuracy of DEM results.

[0145] Step 54: Digital Elevation Model (DEM) Production

[0146] Use the Digital Green Earth LiDAR360 software to output DEM, select the required data, set the resolution and DEM construction, construct a triangulated network for key points, perform interpolation processing, and then transform the grid to raster data to obtain DEM. For details, see Fig.18 and Fig.19 .

[0147] This embodiment performs quality inspection and accuracy evaluation on the two constructed DEMs.

[0148] Specifically, according to the technical requirements of "Airborne LiDAR Point Cloud Data Quality Evaluation Indicators and Calculation Methods" GB / T36100-2018, "Airborne LiDAR Data Acquisition Technical Specifications" CH / T8024-2011, and "Airborne LiDAR Data Processing Technical Specifications" CH / T 8023-2011, point cloud density, elevation accuracy, and digital elevation model accuracy are selected as evaluation indicators of data quality.

[0149] For point cloud density, the point cloud data density is counted separately according to the area. The statistical results are shown in Fig. 20 and Fig.21According to the requirements of the Technical Specifications for Airborne LiDAR Data Acquisition (CH / T 8024-2011), see Table 4 and Table 5 for details. The point cloud data density is greater than 4 / m2, which can meet the 1:1000 scale requirement.

[0150] Table 4 Statistics of point cloud density in each region

[0151] Region Name <![CDATA[Point cloud density (points / m 2 )]]> Remark Before cleaning the East District salt field sedimentation tank 107 After cleaning the East District Salt Field Sedimentation Tank 106

[0152] Table 5 Point cloud density requirements

[0153] scale Digital elevation model grid spacing / meter <![CDATA[Point cloud density (points / m 2 )]]> 1：1000 1.0 ≥4

[0154] For the elevation error, this embodiment uses the QHCORS station to adopt the network RTK method to collect check points before and after the cleaning of the salt field sedimentation pond area in the eastern area; as the elevation error judgment point of the point cloud data, the elevation error statistics are shown in Tables 6 and 7.

[0155] Table 6 Statistics of elevation errors before cleaning of East District salt field sedimentation tank (in meters)

[0156]

[0157]

[0158]

[0159] Table 7 Statistics of elevation errors after cleaning of East District salt field sedimentation tank (in meters)

[0160]

[0161]

[0162]

[0163] Check the process Fig. 22 and Fig.23 .

[0164] According to the requirements of the Technical Specifications for Airborne LiDAR Data Processing (CH / T 8023-2011), see Table 8 for details. The mathematical accuracy of the elevation of the point cloud data obtained this time is better than 1 / 3 of the mean error (terrain category: flat land), meeting the accuracy requirements of the 1:1000 scale.

[0165] Table 8 Point cloud data elevation accuracy requirements

[0166]

[0167] In view of the elevation error of the digital elevation model results, the network RTK method was used to collect check points using the QHCORS station as the elevation error judgment points of the digital elevation model. The elevation error statistics are shown in Tables 9 and 10.

[0168] Table 9 Statistics of elevation errors of the digital elevation model before cleaning of the East District salt field sedimentation tank (in meters)

[0169]

[0170]

[0171]

[0172] Table 10 Statistics of elevation errors of the digital elevation model after cleaning of the East District Salt Field Sedimentation Tank (in meters)

[0173]

[0174]

[0175]

[0176] According to the requirements of the Technical Specifications for Airborne LiDAR Data Acquisition (CH / T 8024-2011), see Table 11 for details. The mathematical accuracy of the digital elevation model acquired this time is better than 1 / 4 mean error (terrain category: flat land), which can meet the 1:1000 scale requirement.

[0177] Table 11 Digital elevation model accuracy requirements

[0178]

[0179] The above implementation modes are only used to illustrate the present application, and are not intended to limit the present application. Ordinary technicians in the relevant technical field may make various changes and modifications without departing from the spirit and scope of the present application. Therefore, all equivalent technical solutions also belong to the scope of the present application, and the scope of patent protection of the present application shall be limited by the claims.

Claims

1. A method for constructing a digital elevation model based on unmanned aerial vehicle laser radar, characterized in that: The method comprises: Based on the UAV equipped with LiDAR payload, fly over the target area to obtain LiDAR point cloud data; Based on the laser radar point cloud data, point cloud data is solved to generate a digital elevation model.

2. The method for constructing a digital elevation model based on a UAV laser radar according to claim 1, characterized in that: Based on the UAV equipped with LiDAR payload, the target area is flown on site to obtain LiDAR point cloud data, including: The goal is to obtain the LiDAR point cloud data of the target area before and after cleaning, and design the flight range and route of the UAV; Determine the take-off and landing site of the drone according to the requirements of the drone equipment operation manual; The UAV is controlled to perform field flight according to a set flight control strategy to obtain laser radar point cloud data.

3. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 2, characterized in that: The flight control strategy is set, including: Flight control method; Aviation and route curvature control methods.

4. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 3, characterized in that: The flight control method includes: During the entire flight, the aircraft's turning slope is controlled within 15°; During the flight along the survey line, the following requirements must be met: Different routes use alternating left and right turns; The deviation from the route does not exceed 25m, and the altitude does not deviate by 25m; The yaw angle is not greater than 6°; The tilt angle of the image is not greater than 2°; The curvature of the route is not greater than 3%; The aircraft speed is maintained at the designed speed for aerial photography.

5. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 3, characterized in that: Aviation and route curvature control methods include: The flight altitude is monitored in real time through the LiDAR measurement system, and corrections are made when the actual flight altitude value deviates from the designed flight altitude.

6. The method for constructing a digital elevation model based on a UAV laser radar according to claim 1, characterized in that: Based on the laser radar point cloud data, point cloud data is solved to generate a digital elevation model, including: Import the laser radar point cloud data into the Qianxunyi laser radar system-point cloud preprocessing module for format conversion, and use the converted IMR data and GNSS data to upload to the Qianxun Cloud Control Console to calculate the trajectory data and radar data; Input the solved trajectory data and radar data into the Qianxunyi LiDAR System-Point Cloud Preprocessing Module, input the engineering path, coordinate system, and set the point cloud generation parameters and filtering parameters in turn to solve the point cloud, and automatically generate high-precision point cloud data; Based on the high-precision point cloud data, the required data is selected, the resolution and DEM construction are set, a triangulated network is constructed for key points, and after interpolation processing, the grid is transformed into raster data to obtain a digital elevation model.

7. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 6, characterized in that: Input the solved trajectory data and radar data into the Qianxunyi LiDAR System-Point Cloud Preprocessing Module, input the engineering path, coordinate system, and set the point cloud generation parameters and filtering parameters in turn to solve the point cloud, and automatically generate high-precision point cloud data, including: Layered inspection; Point cloud denoising; Point cloud post-processing.

8. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 7, characterized in that: Layered inspection, including: The profile and quality inspection tools of the Qianxunyi LiDAR System-Point Cloud Post-Processing Module are used to detect the stratification between routes. If stratification exists, flight strip adjustment is used to reduce the stratification phenomenon.

9. The method for constructing a digital elevation model based on a UAV laser radar as claimed in claim 8, characterized in that: Point cloud denoising, including: removing invalid point clouds and unreasonable point clouds.

10. The method for constructing a digital elevation model based on a UAV laser radar according to claim 9, characterized in that: Point cloud post-processing, including: Point cloud filtering: Construct a triangulated network through the surface point cloud and remove the foot points of the ground objects in the point cloud; classify the ground point cloud based on the iterative morphological filtering method, first divide the survey area into blocks, and the minimum size of the survey area block is larger than the maximum building size in the area, then select the lowest point and the ground points within the set height difference range from the point cloud as ground seed points, establish the initial sparse triangulated network, calculate the angle and distance between the remaining points in the point cloud and the triangulated network, and compare them with the set iteration angle and iteration distance. If it is less than the set iteration value, it is accepted as a ground point, otherwise it is rejected as a ground point, and the above iterations are repeated until no new ground points are added to the triangulated network, thus realizing the process of triangulating network encryption; Point cloud editing and classification: remove gross points that are not filtered out after automatic filtering, identify points that are not correctly classified by automatic classification, and add elevation feature points in areas with terrain distortion.

Images