Scene reconstruction method, device, computer equipment and readable storage medium

Abstract

This application relates to a scene reconstruction method, apparatus, computer device, and computer-readable storage medium. The method comprises: obtaining a point cloud image collected by a vehicle-mounted radar at detection points on a driving trajectory; correcting the point cloud image to be parallel to the ground based on the gravitational acceleration corresponding to the detection points, thereby obtaining a corrected point cloud image; converting the corrected point cloud image to a target coordinate system based on the target heading angle corresponding to the detection points, thereby obtaining a converted point cloud image; and splicing the converted point cloud images to obtain a scene reconstruction result from the vehicle-mounted radar. This method can improve the accuracy of three-dimensional scene reconstruction.

Description

Technical Field

[0001] The present application relates to the field of information technology, and in particular to a scene reconstruction method, apparatus, computer device, and computer-readable storage medium. Background Art

[0002] During the operation and maintenance of power cables, workers may spend a lot of manpower and time to find the cable trajectory. Even if they use equipment such as cable fault detectors, they still cannot locate the fault point quickly and accurately when they know the distance to the fault point but not the cable trajectory.

[0003] To improve the automation of power cable operation and maintenance, relevant detection technologies can be used to obtain data such as the cable route, burial depth, and geographic coordinates of directly buried power cables. This can then be used to construct a spatial model of the power cables and establish a 3D visualization information management and application system for directly buried power cables. For the reconstruction of 3D streetscape information above ground level, a vehicle-mounted LiDAR system can be used to generate high-precision point cloud data by emitting laser pulses and measuring their return time, thereby depicting the 3D structure of streets and buildings.

[0004] However, since the position and posture of the vehicle-mounted lidar system constantly changes as the vehicle moves, it may cause mapping errors, thereby affecting the accuracy of the reconstruction of the three-dimensional street view information. Summary of the Invention

[0005] Based on this, it is necessary to provide a scene reconstruction method, apparatus, computer device, computer-readable storage medium and computer program product that can improve accuracy in order to address the above technical problems.

[0006] In a first aspect, the present application provides a scene reconstruction method, comprising:

[0007] Obtain the point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0008] Correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point to obtain a corrected point cloud image;

[0009] According to the target heading angle corresponding to the detection point, the corrected point cloud image is converted to the target coordinate system to obtain a converted point cloud image;

[0010] The converted point cloud images are spliced ​​to obtain a scene reconstruction result of the vehicle-mounted radar.

[0011] In one embodiment, the step of correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point to obtain the corrected point cloud image includes:

[0012] Determining a first transformation matrix corresponding to the detection point according to the gravitational acceleration;

[0013] The point cloud image collected at the detection point is transformed according to the first transformation matrix to obtain the corrected point cloud image.

[0014] In one embodiment, before converting the corrected point cloud image to a target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud image, the method further includes:

[0015] Determining a plane fitting curve of the driving trajectory;

[0016] Derivative the plane fitting curve at the detection point to obtain an initial heading angle of the detection point;

[0017] The initial heading angle is corrected according to the solar azimuth to obtain the target heading angle corresponding to the detection point.

[0018] In one embodiment, determining the plane fitting curve of the driving trajectory includes:

[0019] Obtaining the latitude and longitude coordinates of the detection point on the driving trajectory;

[0020] Converting the latitude and longitude coordinates into Cartesian coordinates;

[0021] Determining the two-dimensional plane coordinates of the detection point according to the Cartesian coordinates;

[0022] The two-dimensional plane coordinates of each detection point are fitted to obtain the plane fitting curve of the driving trajectory.

[0023] In one embodiment, before correcting the initial heading angle according to the solar azimuth angle to obtain the target heading angle corresponding to the detection point, the method further includes:

[0024] Determining the solar altitude angle according to the latitude coordinate of the detection point;

[0025] The solar azimuth angle corresponding to the detection point is determined according to the solar altitude angle.

[0026] In one embodiment, converting the corrected point cloud image to a target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud image includes:

[0027] Determine a rotation matrix according to the target heading angle corresponding to the detection point, and determine a translation matrix according to the Cartesian coordinates corresponding to the detection point;

[0028] Obtaining a second transformation matrix corresponding to the detection point according to the rotation matrix and the translation matrix;

[0029] The corrected point cloud image is transformed according to the second transformation matrix to obtain the converted point cloud image.

[0030] In a second aspect, the present application further provides a scene reconstruction device, comprising:

[0031] An acquisition module is used to obtain a point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0032] a correction module, configured to correct the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point, thereby obtaining a corrected point cloud image;

[0033] a conversion module, configured to convert the corrected point cloud image into a target coordinate system according to the target heading angle corresponding to the detection point, to obtain a converted point cloud image;

[0034] The splicing module is used to splice the converted point cloud image to obtain the scene reconstruction result of the vehicle-mounted radar.

[0035] In a third aspect, the present application further provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0036] Obtain the point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0037] Correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point to obtain a corrected point cloud image;

[0038] According to the target heading angle corresponding to the detection point, the corrected point cloud image is converted to the target coordinate system to obtain a converted point cloud image;

[0039] The converted point cloud images are spliced ​​to obtain a scene reconstruction result of the vehicle-mounted radar.

[0040] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0041] Obtain the point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0042] Correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point to obtain a corrected point cloud image;

[0043] According to the target heading angle corresponding to the detection point, the corrected point cloud image is converted to the target coordinate system to obtain a converted point cloud image;

[0044] The converted point cloud images are spliced ​​to obtain a scene reconstruction result of the vehicle-mounted radar.

[0045] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which, when executed by a processor, implements the following steps:

[0046] Obtain the point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0047] Correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point to obtain a corrected point cloud image;

[0048] According to the target heading angle corresponding to the detection point, the corrected point cloud image is converted to the target coordinate system to obtain a converted point cloud image;

[0049] The converted point cloud images are spliced ​​to obtain a scene reconstruction result of the vehicle-mounted radar.

[0050] The above-mentioned scene reconstruction method, device, computer equipment, computer-readable storage medium and computer program product obtain the point cloud map collected by the detection points on the driving trajectory of the vehicle-mounted radar, correct the point cloud map to be parallel to the ground according to the gravitational acceleration corresponding to the detection point, and obtain a corrected point cloud map. According to the target heading angle corresponding to the detection point, the corrected point cloud map is converted to the target coordinate system to obtain a converted point cloud map, and the converted point cloud map is spliced ​​to obtain the scene reconstruction result of the vehicle-mounted radar; the point cloud map corresponding to each detection point on the driving trajectory of the vehicle-mounted radar can be first corrected to be parallel to the ground, and then the corrected point cloud map can be converted to the same coordinate system according to the heading angle corresponding to each detection point, so that the converted point cloud map corresponding to each detection point has geographical consistency and spatial continuity, ensuring that a three-dimensional panoramic point cloud map with high accuracy is obtained after splicing. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following briefly introduces the drawings required for use in the embodiments of the present application or related technical descriptions. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying any creative work.

[0052] Figure 1 1 is a flow chart of a scene reconstruction method according to an embodiment;

[0053] Figure 2 A schematic diagram of a flow chart of a method for reconstructing a three-dimensional point cloud scene of a street scene above ground using a directly buried cable in one embodiment;

[0054] Figure 3 is a flow chart of a scene reconstruction method according to another embodiment;

[0055] Figure 4 is a structural block diagram of a scene reconstruction device in one embodiment;

[0056] Figure 5 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0057] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0058] It should be noted that the terms "first," "second," and the like in the specification and claims of the present disclosure and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or precedence. It should be understood that the numbers used in this manner are interchangeable where appropriate so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of apparatus and methods consistent with certain aspects of the present disclosure as detailed in the appended claims.

[0059] In an exemplary embodiment, Figure 1 As shown, a scene reconstruction method is provided. This embodiment uses the method applied to a terminal as an example for illustration. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0060] Step S102: Obtain a point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory.

[0061] The vehicle-mounted radar may be a radar system mounted on a vehicle. The driving trajectory may be the driving trajectory of the vehicle equipped with the radar system. The detection points may be locations along the driving trajectory where the vehicle-mounted radar collects a point cloud image. The point cloud image may be an image generated based on the point cloud data and reflecting three-dimensional information of the scene.

[0062] In the specific implementation, one or more detection points can be set on the driving trajectory. When the vehicle runs to each detection point according to the driving trajectory, the on-board radar can collect point cloud data, generate a point cloud map based on the point cloud data, and send the point cloud map to the terminal. Based on this, the terminal can obtain the point cloud map corresponding to each detection point.

[0063] In practical applications, n detection points can be set on the driving trajectory. When the terminal reaches the i-th (1≤i≤n) detection point, the terminal can obtain the point cloud image collected by the vehicle radar system. In addition, the terminal can also obtain the latitude and longitude coordinates of the vehicle-mounted radar system at the i-th detection point (B i ,L i ,H i ), and the longitude and latitude coordinates (B i ,L i ,H i ) is converted to Cartesian coordinates (x i ,y i ,z i ), where B i ,L i ,H i represent the latitude, longitude, and altitude of the vehicle-mounted radar system at the i-th detection point, respectively, and x i =(N i +H i )cos B i cos L i ,y i =(N i +H i )cosB i sinL i , z i =(N i (1-e 2 )+H i )sin B i ,in is the radius of the Maoyou circle, a is the equatorial radius, b is the polar radius, is the first eccentricity of the Earth.

[0064] It should be noted that the terminal can be a separate device or integrated into a vehicle-mounted radar system.

[0065] In step S104 , the point cloud image is corrected to be parallel to the ground according to the gravity acceleration corresponding to the detection point, thereby obtaining a corrected point cloud image.

[0066] The gravitational acceleration may be the gravitational acceleration when the vehicle-mounted radar moves to the detection point, and the corrected point cloud image may be a point cloud image corrected to be parallel to the ground.

[0067] In the specific implementation, when running to each detection point, the current gravity acceleration of the vehicle-mounted radar can be measured by the gyroscope, and the measured gravity acceleration is sent to the terminal. The terminal determines the transformation matrix of the point cloud map based on the received gravity acceleration, and performs matrix transformation processing on the point cloud map collected at the current detection point according to the transformation matrix to obtain the corrected point cloud map of the current detection point.

[0068] In practical applications, the gyroscope can be built into the vehicle-mounted radar system or rigidly connected to the vehicle-mounted radar system. When it runs to the i-th detection point on the driving trajectory, the gyroscope measures the gravity acceleration of the vehicle-mounted radar system. The terminal can construct a transformation matrix based on gravitational acceleration

[0069]

[0070] in, According to the transformation matrix, the point cloud image collected by the vehicle-mounted radar system at the current detection point is Perform coordinate transformation to obtain the corrected point cloud map The specific formula is

[0071]

[0072] In step S106 , the corrected point cloud image is converted to the target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud image.

[0073] The target heading angle may be the heading angle of the detection point determined according to the solar azimuth. The target coordinate system may be the same coordinate system after the correction of each point cloud image is converted. The converted point cloud image may be a point cloud image that has undergone coordinate transformation.

[0074] In a specific implementation, for each detection point, the terminal can determine the corresponding target heading angle, determine the transformation matrix according to the target heading angle, perform matrix transformation processing on the corresponding corrected point cloud image according to the transformation matrix, and obtain the corresponding converted point cloud image.

[0075] In practical applications, the two-dimensional plane coordinates (x1, y1), (x2, y2), ... (x n ,y n ), use the least square method to find (x1,y1),(x2,y2),…(x n ,y n ) is fitted to obtain the fitting curve y = f(x), and the fitting curve is derived at each detection point, and the inverse tangent function is calculated to obtain the initial heading angle α at each detection point i =arctan(f ′(x)), the initial heading angles of the n detection points can be recorded as α1, α2, …, α n , the error is σ.

[0076] Get the solar azimuth The solar azimuth angles of the n detection points can be recorded as β1, β2, ..., β n , the error is δ, where is the solar altitude angle, is the declination angle, H s is the local solar time, which can be calculated by the time difference and the local standard time. The number of days.

[0077] According to the initial heading angle and the solar azimuth angle, the target heading angle of the i-th detection point can be obtained as follows: The terminal can construct a transformation matrix based on the target heading angle and Cartesian coordinates

[0078]

[0079] According to the transformation matrix, the corrected point cloud image Perform coordinate transformation to obtain the converted point cloud map The specific formula is

[0080]

[0081] Step S108: stitching the converted point cloud images to obtain the scene reconstruction result of the vehicle-mounted radar.

[0082] The scene reconstruction result may be a panoramic point cloud image of a reconstructed three-dimensional scene.

[0083] In a specific implementation, the terminal can stitch together the converted point cloud images corresponding to each detection point to obtain a point cloud image for reconstructing the three-dimensional scene, and use the point cloud image as the scene reconstruction result of the vehicle-mounted radar.

[0084] In practical applications, the converted point cloud corresponding to each detection point can be Stitched together, we can get the panoramic cloud map of the reconstructed three-dimensional scene.

[0085] The above-mentioned scene reconstruction method obtains the point cloud map collected by the detection points on the driving trajectory of the vehicle-mounted radar, corrects the point cloud map to be parallel to the ground according to the gravitational acceleration corresponding to the detection point, and obtains the corrected point cloud map. According to the target heading angle corresponding to the detection point, the corrected point cloud map is converted to the target coordinate system to obtain the converted point cloud map, and the converted point cloud map is spliced ​​to obtain the scene reconstruction result of the vehicle-mounted radar; the point cloud map corresponding to each detection point on the driving trajectory of the vehicle-mounted radar can be first corrected to be parallel to the ground, and then the corrected point cloud map can be converted to the same coordinate system according to the heading angle corresponding to each detection point, so that the converted point cloud map corresponding to each detection point has geographical consistency and spatial continuity, ensuring that a three-dimensional panoramic point cloud map with high accuracy is obtained after splicing.

[0086] In an exemplary embodiment, the above step S104 may specifically include: determining a first transformation matrix corresponding to the detection point according to the gravitational acceleration; and transforming the point cloud image collected at the detection point according to the first transformation matrix to obtain a corrected point cloud image.

[0087] The first transformation matrix may be a matrix for correcting the elevation angle of the point cloud image.

[0088] In a specific implementation, the point cloud image collected by the vehicle-mounted radar may have an elevation angle. In order to stitch the point cloud images collected at each detection point together, the elevation angles of each point cloud image can be corrected to 0 so that the XY plane of the point cloud image is parallel to the ground. Specifically, at the i-th detection point, the gravity acceleration of the vehicle-mounted radar can be measured by the gyroscope. Terminal according to Determine the first transformation matrix corresponding to the i-th detection point

[0089]

[0090] in, According to the first transformation matrix, the point cloud image collected at the i-th detection point Perform coordinate transformation to obtain the corrected point cloud map The specific formula is

[0091]

[0092] In this embodiment, the first transformation matrix corresponding to the detection point is determined based on the gravitational acceleration, and the point cloud map collected at the detection point is transformed according to the first transformation matrix to obtain a corrected point cloud map. The point cloud maps collected at each detection point can be corrected to be parallel to the ground, so that the elevation angles of each point cloud map are consistent, providing convenient conditions for subsequent splicing.

[0093] In an exemplary embodiment, before step S106, the process may further include: determining a plane fitting curve of the driving trajectory; taking the derivative of the plane fitting curve at the detection point to obtain an initial heading angle of the detection point; and correcting the initial heading angle according to the solar azimuth to obtain a target heading angle corresponding to the detection point.

[0094] The plane fitting curve may be a fitting curve of the driving trajectory on a two-dimensional plane, and the initial heading angle may be a heading angle of the detection point determined according to the plane fitting curve of the driving trajectory.

[0095] In a specific implementation, for n detection points on the driving trajectory, the terminal can obtain the Cartesian coordinates (x i ,y i ,z i ),1≤i≤n, select the two-dimensional plane coordinates (x1,y1),(x2,y2),…(x n ,y n ), use the least square method to find (x1,y1),(x2,y2),…(x n ,y n ) is fitted to obtain the plane fitting curve y = f(x). The plane fitting curve is derived at each detection point, and the inverse tangent function is calculated to obtain the initial heading angle α corresponding to each detection point. i =arctan(f ′ (x)), 1≤i≤n, the error is σ. The terminal can also calculate the solar azimuth angle corresponding to each detection point The error is δ. The initial heading angle is corrected using the solar azimuth angle to obtain the target heading angle corresponding to each detection point.

[0096] In this embodiment, by determining the plane fitting curve of the driving trajectory, the plane fitting curve is differentiated at the detection point to obtain the initial heading angle of the detection point, and the initial heading angle is corrected according to the solar azimuth to obtain the target heading angle corresponding to the detection point. The heading angle of each detection point can be determined according to the solar azimuth, avoiding the inaccuracy caused by the gyroscope in calculating the heading angle, and improving the accuracy of scene reconstruction.

[0097] In an exemplary embodiment, the above-mentioned step of determining the plane fitting curve of the driving trajectory may specifically include: obtaining the latitude and longitude coordinates of the detection points on the driving trajectory; converting the latitude and longitude coordinates into Cartesian coordinates; determining the two-dimensional plane coordinates of the detection points based on the Cartesian coordinates; fitting the two-dimensional plane coordinates of each detection point to obtain the plane fitting curve of the driving trajectory.

[0098] In the specific implementation, the terminal can obtain the latitude and longitude coordinates of each detection point on the vehicle radar driving trajectory (B i ,Li ,H i ), and the longitude and latitude coordinates (B i ,L i ,H i ) is converted to Cartesian coordinates (x i ,y i ,z i ), where x i =(N i +H i )cosB i cosL i ,y i =(N i +H i )cos B i sinL i , z i =(N i (1-e 2 )+H i )sibB i ,in is the radius of the Maoyou circle, a is the equatorial radius, b is the polar radius, is the first eccentricity of the Earth. The terminal can also be converted from Cartesian coordinates (x i ,y i ,z i ) in the two-dimensional plane coordinates (x i ,y i ), and obtain the two-dimensional plane coordinates of each detection point (x1, y1), (x2, y2), ... (x n ,y n ), use the least square method to find (x1,y1),(x2,y2),…(x n ,y n ) is fitted to obtain the plane fitting curve y=f(x) of the vehicle-mounted radar driving trajectory.

[0099] In this embodiment, the latitude and longitude coordinates of the detection points on the driving trajectory are obtained, the latitude and longitude coordinates are converted into Cartesian coordinates, and the two-dimensional plane coordinates of the detection points are determined based on the Cartesian coordinates. The two-dimensional plane coordinates of each detection point are fitted to obtain a plane fitting curve of the driving trajectory. The plane fitting curve can be extracted from the three-dimensional driving trajectory of the vehicle-mounted radar, providing convenient conditions for determining the heading angle based on the plane fitting curve, thereby improving the accuracy of scene reconstruction.

[0100] In an exemplary embodiment, before the above-mentioned step of correcting the initial heading angle according to the solar azimuth angle to obtain the target heading angle corresponding to the detection point, it can also specifically include: determining the solar altitude angle according to the latitude coordinate of the detection point; and determining the solar azimuth angle corresponding to the detection point according to the solar altitude angle.

[0101] In the specific implementation, the terminal can use the latitude coordinate B of the i-th detection point i , get the solar altitude angle in is the declination angle, H s is the local solar time, The terminal can also calculate the number of days according to the solar altitude angle h i , get the solar azimuth angle corresponding to the i-th detection point

[0102] In this embodiment, by determining the solar altitude angle according to the latitude coordinate of the detection point, and determining the solar azimuth angle corresponding to the detection point according to the solar altitude angle, the solar azimuth angle can be quickly determined, and the heading angle can be determined using the solar azimuth angle, thereby improving the accuracy of scene reconstruction.

[0103] In an exemplary embodiment, the above-mentioned step S106 may specifically include: determining a rotation matrix based on the target heading angle corresponding to the detection point, and determining a translation matrix based on the Cartesian coordinates corresponding to the detection point; obtaining a second transformation matrix corresponding to the detection point based on the rotation matrix and the translation matrix; and transforming the corrected point cloud image based on the second transformation matrix to obtain a converted point cloud image.

[0104] The second transformation matrix may be a matrix for rotating and translating the corrected point cloud image parallel to the ground.

[0105] In the specific implementation, for the i-th detection point, the terminal can Constructing the rotation matrix According to the Cartesian coordinates (x i ,y i ,z i )Construct the translation matrix Construct the second transformation matrix based on the rotation matrix and translation matrix

[0106]

[0107] According to T, the corrected point cloud corresponding to the i-th detection point Perform coordinate transformation to obtain the transformed point cloud corresponding to the i-th detection point The specific formula is

[0108]

[0109] In this embodiment, the rotation matrix is ​​determined according to the target heading angle corresponding to the detection point, the translation matrix is ​​determined according to the Cartesian coordinates corresponding to the detection point, and the second transformation matrix corresponding to the detection point is obtained according to the rotation matrix and the translation matrix. According to the second transformation matrix, the corrected point cloud map is transformed to obtain the converted point cloud map. The corrected point cloud map parallel to the ground can be converted to the same coordinate system to ensure that a scene reconstructed point cloud map with higher accuracy is obtained after splicing.

[0110] In order to facilitate those skilled in the art to have a deeper understanding of the embodiments of the present application, a specific example will be used for illustration below.

[0111] During the 3D reconstruction of streetscape information, vehicle-mounted LiDAR systems generate high-precision point cloud data by emitting laser pulses and measuring their return times. This point cloud data is then used to depict the 3D structure of streets and buildings. Feature extraction and matching are then performed, pairing feature points from multiple perspectives. These paired points are then used for accurate 3D reconstruction, ultimately generating a detailed 3D model of the target streetscape area. However, because the position and attitude of the vehicle-mounted LiDAR system constantly change as the vehicle moves, excessive gyroscope vibration can lead to inaccurate data collection, resulting in mapping errors. Furthermore, vehicle movement can cause temporal and spatial inconsistencies in LiDAR data collection, making it difficult to effectively fuse and generate a complete 3D panoramic model. It can even prevent the effective splicing of 3D data from multiple data collection points, resulting in a loss of 3D panoramic model construction. Therefore, a high-precision synchronization and registration algorithm is currently needed to fuse this data to ensure that the resulting 3D model possesses accurate geographic consistency and spatial continuity.

[0112] This application provides a method for reconstructing a 3D point cloud scene of a directly buried cable above ground street view. This method uses the positioning data during the movement and the azimuth of the sun to stitch the point cloud images collected by the radar system at different times to form a complete 3D street view cloud. Figure 2 , the method may include the following steps:

[0113] In step S1, the device is placed at the initial point, and the latitude and longitude coordinates of the current point (B1, L1, H1) are obtained, along with the point cloud of the current point. The device is placed at the next location, and the latitude and longitude coordinates of the current point (B2, L2, H2) are obtained, along with the point cloud of the LiDAR at the current point. This step is repeated until the entire line inspection process is completed.

[0114] The point cloud image of the laser radar at the current point is a point cloud image based on the radar coordinate system. The coordinates of the point cloud are in is the coordinate of point i in the radar coordinate system at location i.

[0115] Step S2: The longitude and latitude coordinates (B1, L1, H1), (B2, L2, H2), ... (B n ,L n ,H n ) is converted into a Cartesian coordinate system. The converted coordinates are (x1, y1, z1), (x2, y2, z2),… (x n ,y n ,z n ). x=(N+H)cosBcosL, y=(N+H)cosBsinL, z=(N(1-e 2 )+H)sinB. Where a is the equatorial radius, b is the polar radius, and N is the circumpolar radius. e is the first eccentricity of the Earth

[0116] Step S3, the point cloud of each position point is "corrected" by the built-in gyroscope of the laser radar. In the actual process, the device point cloud will have an elevation angle. Through "correction", the XY plane of the laser radar point cloud is made parallel to the ground. The laser radar has a built-in six-axis gyroscope, which uses gravity acceleration to calculate and transform the laser radar point cloud to be parallel to the ground, and then store the point cloud map of each position point. The coordinates of the laser radar point i at the i-th position are The acceleration due to gravity is The coordinates of the converted point are

[0117]

[0118] in

[0119] Step S4, using the least squares method to calculate the coordinates of each point in the process of equipment patrol (x1, y1), (x2, y2), ... (x n ,y n ) is fitted into a curve, and the fitted curve is y=f(x).

[0120] Step S5, derive each position Get the heading angle α at each position i =arctan(f ′ (x)), the angles are α1, α2, …, α n , the error is σ.

[0121] Step S6, obtaining the azimuth of the sun The azimuth angles of each point in the trajectory are β1,β2,…,β n , the error is δ. H sIt is the local solar time, which can be calculated by the time difference with the local standard time.

[0122] Step S7, calculate the heading angle of each point as

[0123] Step S8, convert the point cloud of each point into the same coordinate system to generate a point cloud map of the entire line inspection process. The conversion formula for converting the point cloud at location i into the same coordinate system is:

[0124]

[0125] This application proposes an algorithm for calculating heading angles using the solar azimuth and fitted trajectories, compensating for the inaccuracy of gyroscopes in calculating heading angles and enabling the construction of three-dimensional streetscapes. Specifically, this application uses a mapping method based on trajectories and solar azimuths to avoid the cumulative error of gyroscope heading angles during the mapping process, improving the stability and accuracy of the mapping. Furthermore, this application uses data from different moments, eliminating the need for spatial and temporal unification, reducing the computing power required for on-site processing on the device and making it suitable for different scenarios.

[0126] In one embodiment, Figure 3 As shown, a scene reconstruction method is provided, which is described by taking the application of the method to a terminal as an example, and includes the following steps:

[0127] Step S301, obtaining a point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0128] Step S302: Correcting the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point, thereby obtaining a corrected point cloud image;

[0129] Step S303: obtaining the longitude and latitude coordinates of the detection points, converting the longitude and latitude coordinates into Cartesian coordinates, determining the two-dimensional plane coordinates of the detection points based on the Cartesian coordinates, and fitting the two-dimensional plane coordinates of each detection point to obtain a plane fitting curve of the driving trajectory;

[0130] Step S304: Derivative the plane fitting curve at the detection point to obtain an initial heading angle of the detection point, and correct the initial heading angle according to the solar azimuth to obtain a target heading angle corresponding to the detection point;

[0131] Step S305: converting the corrected point cloud image to the target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud image;

[0132] Step S306: stitching the converted point cloud images to obtain the scene reconstruction result of the vehicle-mounted radar.

[0133] In a specific implementation, the terminal can obtain the point cloud map collected by the on-board radar at each detection point on the driving trajectory, construct a transformation matrix based on the gravitational acceleration corresponding to each detection point, and use the transformation matrix to correct the point cloud map to be parallel to the ground, thereby obtaining the corrected point cloud map corresponding to each detection point. The terminal can also obtain the latitude and longitude coordinates of the on-board radar at each detection point, convert the latitude and longitude coordinates into Cartesian coordinates, extract two-dimensional plane coordinates from the Cartesian coordinates, fit the two-dimensional plane coordinates corresponding to each detection point, obtain a plane fitting curve of the driving trajectory, and successively perform differentiation and inverse tangent function on the plane fitting curve to obtain the initial heading angle corresponding to each detection point. The initial heading angle is corrected according to the solar azimuth angle to obtain the target heading angle corresponding to each detection point. The target heading angle is used to construct a transformation matrix, and the corrected point cloud map corresponding to each detection point is converted to the same coordinate system, i.e., the target coordinate system, through the transformation matrix to obtain a converted point cloud map. The converted point cloud maps corresponding to each detection point are spliced ​​together to obtain a panoramic point cloud map for three-dimensional scene reconstruction.

[0134] The above scene reconstruction method obtains the point cloud map collected by the vehicle-mounted radar at the detection point on the driving trajectory, corrects the point cloud map to be parallel to the ground according to the gravity acceleration corresponding to the detection point, obtains the latitude and longitude coordinates of the detection point, converts the latitude and longitude coordinates into Cartesian coordinates, determines the two-dimensional plane coordinates of the detection point according to the Cartesian coordinates, fits the two-dimensional plane coordinates of each detection point, obtains the plane fitting curve of the driving trajectory, derivates the plane fitting curve at the detection point, obtains the initial heading angle of the detection point, corrects the initial heading angle according to the solar azimuth, and obtains The target heading angle corresponding to the detection point is used to convert the corrected point cloud map to the target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud map, and the converted point cloud map is spliced ​​to obtain the scene reconstruction result of the vehicle-mounted radar; the point cloud map corresponding to each detection point on the driving trajectory of the vehicle-mounted radar can be first corrected to be parallel to the ground, and then the corrected point cloud map can be converted to the same coordinate system according to the heading angle corresponding to each detection point, so that the converted point cloud map corresponding to each detection point has geographical consistency and spatial continuity, ensuring that a three-dimensional panoramic point cloud map with high accuracy is obtained after splicing.

[0135] It should be understood that, although the various steps in the flowcharts involved in the various embodiments described above are displayed in sequence according to the instructions of the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and these steps can be executed in other orders. Moreover, at least a portion of the steps in the flowcharts involved in the various embodiments described above can include multiple steps or multiple stages, and these steps or stages are not necessarily executed and completed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a portion of steps or stages in other steps.

[0136] Based on the same inventive concept, embodiments of the present application also provide a scene reconstruction device for implementing the scene reconstruction method described above. The solution provided by this device is similar to the solution described in the method described above. Therefore, the specific limitations of one or more of the following scene reconstruction device embodiments can be found in the above-mentioned limitations of the scene reconstruction method and will not be repeated here.

[0137] In an exemplary embodiment, Figure 4 As shown, a scene reconstruction device is provided, including: an acquisition module 402, a correction module 404, a conversion module 406 and a splicing module 408, wherein:

[0138] An acquisition module 402 is used to acquire a point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory;

[0139] A correction module 404 is configured to correct the point cloud image to be parallel to the ground according to the gravity acceleration corresponding to the detection point, thereby obtaining a corrected point cloud image;

[0140] A conversion module 406 is configured to convert the corrected point cloud image into a target coordinate system according to the target heading angle corresponding to the detection point, thereby obtaining a converted point cloud image;

[0141] The stitching module 408 is used to stitch the converted point cloud image to obtain the scene reconstruction result of the vehicle-mounted radar.

[0142] In an exemplary embodiment, the correction module 404 is further used to determine a first transformation matrix corresponding to the detection point based on the gravitational acceleration; and to perform transformation processing on the point cloud image collected at the detection point based on the first transformation matrix to obtain the corrected point cloud image.

[0143] In an exemplary embodiment, the above-mentioned scene reconstruction device also includes a calculation module for determining a plane fitting curve of the driving trajectory; derivatizing the plane fitting curve at the detection point to obtain the initial heading angle of the detection point; and correcting the initial heading angle according to the solar azimuth angle to obtain the target heading angle corresponding to the detection point.

[0144] In an exemplary embodiment, the above-mentioned calculation module is also used to obtain the latitude and longitude coordinates of the detection points on the driving trajectory; convert the latitude and longitude coordinates into Cartesian coordinates; determine the two-dimensional plane coordinates of the detection points based on the Cartesian coordinates; and fit the two-dimensional plane coordinates of each of the detection points to obtain the plane fitting curve of the driving trajectory.

[0145] In an exemplary embodiment, the calculation module is further configured to determine the solar altitude angle according to the latitude coordinate of the detection point; and determine the solar azimuth angle corresponding to the detection point according to the solar altitude angle.

[0146] In an exemplary embodiment, the above-mentioned conversion module 406 is also used to determine a rotation matrix based on the target heading angle corresponding to the detection point, and determine a translation matrix based on the Cartesian coordinates corresponding to the detection point; obtain a second transformation matrix corresponding to the detection point based on the rotation matrix and the translation matrix; and transform the corrected point cloud image based on the second transformation matrix to obtain the converted point cloud image.

[0147] Each module in the scene reconstruction device described above can be implemented in whole or in part through software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in the computer device in the form of software, so that the processor can call and execute the corresponding operations of each module.

[0148] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 5As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected via a system bus, and the communication interface, the display unit and the input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and an external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, near field communication (NFC) or other technologies. When the computer program is executed by the processor, a scene reconstruction method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0149] Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0150] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0151] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0152] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0153] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0154] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiment methods can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, database or other media used in the embodiments provided in this application may include at least one of non-volatile memory and volatile memory. Non-volatile memory may include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory may include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processor involved in the various embodiments provided herein may be, but are not limited to, a general-purpose processor, a central processing unit (CPU), a graphics processing unit (GPU), a digital signal processor (DSP), a programmable logic unit (PLC), a data processing logic unit based on quantum computing, an artificial intelligence (AI) processor, and the like.

[0155] The technical features of the above embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0156] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A scene reconstruction method, characterized in that: The method comprises: Obtain the point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory; determining a first transformation matrix corresponding to the detection point according to the gravitational acceleration corresponding to the detection point, correcting the elevation angle of the point cloud image according to the first transformation matrix, correcting the point cloud image to be parallel to the ground, and obtaining a corrected point cloud image; Determining a plane fitting curve of the driving trajectory, derivatizing the plane fitting curve at the detection point to obtain an initial heading angle of the detection point, and correcting the initial heading angle according to a solar azimuth to obtain a target heading angle corresponding to the detection point; According to the target heading angle corresponding to the detection point, the corrected point cloud image is converted to a target coordinate system to obtain a converted point cloud image; The converted point cloud images are spliced ​​to obtain a scene reconstruction result of the vehicle-mounted radar.

2. The method according to claim 1, characterized in that The determining of the plane fitting curve of the driving trajectory includes: Obtaining the latitude and longitude coordinates of the detection point on the driving trajectory; Converting the latitude and longitude coordinates into Cartesian coordinates; Determining the two-dimensional plane coordinates of the detection point according to the Cartesian coordinates; The two-dimensional plane coordinates of each detection point are fitted to obtain the plane fitting curve of the driving trajectory.

3. The method according to claim 1, characterized in that Before correcting the initial heading angle according to the solar azimuth angle to obtain the target heading angle corresponding to the detection point, the method further includes: Determining the solar altitude angle according to the latitude coordinate of the detection point; The solar azimuth angle corresponding to the detection point is determined according to the solar altitude angle.

4. The method according to claim 1, wherein The step of converting the corrected point cloud image to a target coordinate system according to the target heading angle corresponding to the detection point to obtain a converted point cloud image includes: Determine a rotation matrix according to the target heading angle corresponding to the detection point, and determine a translation matrix according to the Cartesian coordinates corresponding to the detection point; Obtaining a second transformation matrix corresponding to the detection point according to the rotation matrix and the translation matrix; The corrected point cloud image is transformed according to the second transformation matrix to obtain the converted point cloud image.

5. A scene reconstruction device, characterized in that: The device comprises: An acquisition module is used to obtain a point cloud image collected by the vehicle-mounted radar at the detection points on the driving trajectory; a correction module, configured to determine a first transformation matrix corresponding to the detection point according to the gravitational acceleration corresponding to the detection point, correct the elevation angle of the point cloud image according to the first transformation matrix, correct the point cloud image to be parallel to the ground, and obtain a corrected point cloud image; a calculation module, configured to determine a plane fitting curve of the driving trajectory, derive the plane fitting curve at the detection point to obtain an initial heading angle of the detection point, and correct the initial heading angle according to a solar azimuth to obtain a target heading angle corresponding to the detection point; a conversion module, configured to convert the corrected point cloud image into a target coordinate system according to the target heading angle corresponding to the detection point, to obtain a converted point cloud image; The splicing module is used to splice the converted point cloud image to obtain the scene reconstruction result of the vehicle-mounted radar.

6. The device according to claim 5, characterized in that The calculation module is also used to obtain the longitude and latitude coordinates of the detection points on the driving trajectory, convert the longitude and latitude coordinates into Cartesian coordinates, determine the two-dimensional plane coordinates of the detection points based on the Cartesian coordinates, and fit the two-dimensional plane coordinates of each detection point to obtain the plane fitting curve of the driving trajectory.

7. The device according to claim 5, characterized in that The calculation module is further used to determine the solar altitude angle according to the latitude coordinate of the detection point, and determine the solar azimuth angle corresponding to the detection point according to the solar altitude angle.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 4 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

10. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 4 are implemented.

Images