Elevation information generation method, apparatus, device, and medium

Abstract

The present application relates to the field of map data processing, and discloses an elevation information generation method, device, equipment and medium, the method comprises: obtaining crowd-sourced trajectories, and extracting feature point information of the crowd-sourced trajectories; according to the feature point information, layer height calculation is carried out, corresponding interlayer height value is determined, and the crowd-sourced trajectories are stretched according to the interlayer height value, height dimension information is determined; the crowd-sourced trajectories are split to corresponding plane maps, and the crowd-sourced trajectory segments corresponding to each road section of the plane map are obtained; the crowd-sourced trajectory segments are analyzed by principal component analysis (PCA), the main direction of the crowd-sourced trajectory segments is determined, and corresponding elevation information is generated according to the height dimension information and the main direction. The present application determines the elevation information based on the crowd-sourced trajectories, reduces the equipment cost in the data collection process, solves the problem of inconsistent data quality of the collected data, reduces the data deviation, and improves the precision of the elevation information.

Description

Technical Field

[0001] The present invention relates to the field of map data processing, and in particular to a method, device, equipment and medium for generating elevation information. Background Art

[0002] High-precision maps are an indispensable mapping tool for unmanned vehicles to achieve autonomous driving. With the three-dimensional development of high-precision maps, obtaining altitude information has become an important part of building high-precision maps. Existing solutions often collect altitude information in maps through various sensor devices, such as IMU, odometer, and visual acquisition, which increases the equipment cost of obtaining altitude information. In addition, the data collection process requires that the installation and configuration of each sensor be consistent. When the data quality obtained by each collection device is inconsistent, the estimated elevation information has large deviations and low accuracy.

[0003] Therefore, it is necessary to propose a solution to improve the accuracy of elevation information. Summary of the Invention

[0004] The main purpose of the present invention is to provide a method, device, equipment and medium for generating elevation information, aiming to improve the accuracy of elevation information.

[0005] To achieve the above object, the present invention further provides a method for generating elevation information, the method comprising:

[0006] Obtaining a crowdsourcing trajectory and extracting feature point information of the crowdsourcing trajectory;

[0007] Calculating the layer height according to the feature point information to determine the corresponding inter-layer height value, and stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information;

[0008] Splitting the crowdsourced trajectory into corresponding planar maps, and obtaining crowdsourced trajectory segments corresponding to each road section of the planar map;

[0009] A PCA direction analysis is performed on the crowdsourced trajectory segment to determine the main direction of the crowdsourced trajectory segment, and corresponding elevation information is generated according to the height dimension information and the main direction.

[0010] Optionally, the step of extracting feature point information of the crowdsourcing trajectory includes:

[0011] According to the crowdsourced trajectory, obtaining single trajectory data corresponding to the single trajectory;

[0012] According to the height change information of the single track data, a height feature point is determined, and feature point information of the height feature point is extracted.

[0013] Optionally, the step of calculating the layer height according to the feature point information to determine the corresponding inter-layer height value includes:

[0014] Obtaining feature point information of several single trajectories of the crowdsourced trajectory;

[0015] Performing statistics on the feature point information corresponding to the plurality of single tracks to determine the inter-layer heights of the plurality of single tracks;

[0016] The inter-layer heights of the plurality of single trajectories are calculated to determine an inter-layer height value of the crowdsourced trajectory.

[0017] Optionally, the step of stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information includes:

[0018] Substituting the interlayer height value into a preset formula for calculation to determine the interlayer stretching factor;

[0019] performing a height dimension stretching calculation on the single track data corresponding to the single track according to the inter-layer stretching factor to determine the height dimension information of the single track;

[0020] Based on the height dimension information of the single trajectory, the height dimension information of the crowdsourced trajectory is determined.

[0021] Optionally, the step of performing principal component analysis (PCA) on the crowdsourcing trajectory segments to determine the main directions of the crowdsourcing trajectory segments includes:

[0022] Performing a PCA analysis on each single trajectory segment in the crowdsourced trajectory segmentation to determine the trajectory direction corresponding to each single trajectory segment;

[0023] A PCA secondary analysis is performed on the trajectory direction corresponding to each single trajectory segment in the crowdsourcing trajectory to determine the main direction of the crowdsourcing trajectory segment.

[0024] Optionally, after the step of determining the main direction of the road section where the crowdsourced trajectory segment is located, the method further includes:

[0025] transforming the trajectory direction of each single trajectory segment in the crowdsourced trajectory segment to the main direction, and determining the main direction of the road segment where the crowdsourced trajectory segment is located;

[0026] The step of generating corresponding elevation information according to the height dimension information and the main direction comprises:

[0027] Obtaining height dimension information of the road section where the crowdsourced trajectory segment is located;

[0028] Denoising and fusing the main direction of the road section where the crowdsourced trajectory segment is located and the height dimension information of the road section where the crowdsourced trajectory segment is located are performed to determine the elevation information of the road section where the crowdsourced trajectory segment is located.

[0029] Optionally, after the step of obtaining the crowdsourcing trajectory, the method further includes:

[0030] Matching the ground information of the crowd-sourced trajectory to determine the plane map corresponding to the crowd-sourced trajectory;

[0031] If the crowdsourced track is located in the intersection area of ​​the planar map, obtaining the intersection adjacent section corresponding to the crowdsourced track in the planar map;

[0032] Acquire the height information of the adjacent road sections of the intersection, perform floor height calculation and trajectory stretching on the height information of the adjacent road sections of the intersection, and determine the height dimension information corresponding to the intersection area in the plane map.

[0033] To achieve the above object, the present invention further provides an elevation information generating device, comprising:

[0034] An acquisition module, configured to acquire a crowdsourcing trajectory and extract feature point information of the crowdsourcing trajectory;

[0035] a calculation module, configured to calculate the layer height according to the feature point information, determine the corresponding inter-layer height value, and stretch the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information;

[0036] A splitting module is used to split the crowdsourced trajectory into corresponding plane maps and obtain crowdsourced trajectory segments corresponding to each road section of the plane map;

[0037] The analysis module is used to perform PCA direction analysis on the crowd-sourced trajectory segment, determine the main direction of the crowd-sourced trajectory segment, and generate corresponding elevation information based on the height dimension information and the main direction.

[0038] Among them, each functional module of the elevation information generating device of the present invention implements the steps of the elevation information generating method described above when running.

[0039] To achieve the above-mentioned purpose, the present invention also provides a device, which includes: a memory, a processor, and an elevation information generation program stored in the memory and runnable on the processor. When the elevation information generation program is executed by the processor, the steps of the elevation information generation method described above are implemented.

[0040] To achieve the above objectives, the present invention also proposes a computer-readable storage medium, on which an elevation information generation program is stored. When the elevation information generation program is executed by a processor, the steps of the elevation information generation method described above are implemented.

[0041] The present invention provides a method, device, equipment and medium for generating elevation information. The method includes: obtaining a crowdsourced trajectory and extracting feature point information of the crowdsourced trajectory; calculating layer heights based on the feature point information to determine corresponding inter-layer height values, and stretching the crowdsourced trajectory based on the inter-layer height values ​​to determine elevation dimension information; splitting the crowdsourced trajectory into corresponding plane maps to obtain crowdsourced trajectory segments corresponding to each road section of the plane map; performing principal component analysis (PCA) on the crowdsourced trajectory segments to determine the main directions of the crowdsourced trajectory segments, and generating corresponding elevation information based on the elevation dimension information and the main directions.

[0042] Compared to the current method of collecting elevation information from driving trajectory data by equipping vehicles or other collection equipment with devices such as lidar, sensors, IMUs, and odometers, this solution obtains crowdsourced trajectories and performs floor height calculations and trajectory stretching based on them to determine the corresponding elevation dimension information. PCA analysis is then performed on the crowdsourced trajectories to determine their principal directions. Finally, elevation information is generated from this elevation dimension and principal direction. This approach, which determines elevation information based on crowdsourced trajectories, reduces reliance on high-precision detection equipment and lowers equipment costs during the data collection process. By directly calculating trajectory data, this solves the problem of inconsistent collected data quality, reduces data deviations during subsequent calculations and inference, and improves the accuracy of elevation information. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic diagram of the hardware operating environment involved in an embodiment of the present invention;

[0044] Figure 2 1 is a flow chart of a first embodiment of a method for generating elevation information according to the present invention;

[0045] Figure 3 1 is a flow chart of a second embodiment of the method for generating elevation information according to the present invention;

[0046] Figure 4 This is a sub-flow diagram of step S10 in the second embodiment of the elevation information generation method of the present invention;

[0047] Figure 5 A schematic diagram of a sub-flow diagram of the third embodiment of the elevation information generating method of the present invention;

[0048] Figure 6This is another sub-flow diagram of the third embodiment of the elevation information generating method of the present invention;

[0049] Figure 7 A schematic diagram of a specific flow chart of an exemplary embodiment of the third embodiment of the elevation information generating method of the present invention;

[0050] Figure 8 1 is a flow chart of a fourth embodiment of a method for generating elevation information according to the present invention;

[0051] Figure 9 This is a schematic diagram of a specific flow chart of an exemplary embodiment of the fourth embodiment of the elevation information generating method of the present invention;

[0052] Figure 10 This is a schematic diagram of the functional modules of the elevation information generating device involved in the elevation information generating method of the present invention.

[0053] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0054] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0055] Explanation of the terminology of the present invention:

[0056] Principal components analysis (PCA) aims to transform multiple indicators into a few comprehensive metrics using the concept of dimensionality reduction. PCA analysis is a technique for simplifying data sets by switching the data into a new coordinate system through linear transformation, so that the largest variance of any data projection is on the first coordinate (called the first principal component), the second largest variance is on the second coordinate (the second principal component), and so on. Principal component analysis can reduce the dimensionality of a data set while retaining the features that contribute most to the variance, preserving the most important aspects of the data.

[0057] Specifically, refer to Figure 1 , Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiment of the present invention. Figure 1As shown, the device may include: a processor 1001, such as a CPU, a network interface 1004, a user interface 1003, a memory 1005, and a communication bus 1002. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface). The memory 1005 may be a high-speed RAM memory, or a stable memory (non-volatile memory), such as a disk memory. The memory 1005 may optionally be a storage device independent of the aforementioned processor 1001.

[0058] like Figure 1 As shown, memory 1005, a computer storage medium, may include an operating system, a network communication module, a user interface module, and an elevation information generation program. The operating system is a program that manages and controls device hardware and software resources, supporting the operation of the elevation information generation program and other software or programs. The network communication module is used to manage and control the network interface 1002. The user interface 1003 is primarily used for data communication with the client. The network interface 1004 is primarily used to establish a communication connection with the server. The processor 1001 can be used to call the elevation information generation program stored in memory 1005.

[0059] When the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are implemented:

[0060] Obtaining a crowdsourcing trajectory and extracting feature point information of the crowdsourcing trajectory;

[0061] Calculating the layer height according to the feature point information to determine the corresponding inter-layer height value, and stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information;

[0062] Splitting the crowdsourced trajectory into corresponding planar maps, and obtaining crowdsourced trajectory segments corresponding to each road section of the planar map;

[0063] A principal component analysis (PCA) is performed on the crowdsourced trajectory segment to determine the main direction of the crowdsourced trajectory segment, and corresponding elevation information is generated according to the height dimension information and the main direction.

[0064] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0065] According to the crowdsourced trajectory, obtaining single trajectory data corresponding to the single trajectory;

[0066] According to the height change information of the single track data, a height feature point is determined, and feature point information of the height feature point is extracted.

[0067] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0068] Obtaining feature point information of several single trajectories of the crowdsourced trajectory;

[0069] Performing statistics on the feature point information corresponding to the plurality of single tracks to determine the inter-layer heights of the plurality of single tracks;

[0070] The inter-layer heights of the plurality of single trajectories are calculated to determine an inter-layer height value of the crowdsourced trajectory.

[0071] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0072] Substituting the interlayer height value into a preset formula for calculation to determine the interlayer stretching factor;

[0073] performing a height dimension stretching calculation on the single track data corresponding to the single track according to the inter-layer stretching factor to determine the height dimension information of the single track;

[0074] Based on the height dimension information of the single trajectory, the height dimension information of the crowdsourced trajectory is determined.

[0075] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0076] Performing a PCA analysis on each single trajectory segment in the crowdsourced trajectory segmentation to determine the trajectory direction corresponding to each single trajectory segment;

[0077] A PCA secondary analysis is performed on the trajectory direction corresponding to each single trajectory segment in the crowdsourcing trajectory to determine the main direction of the crowdsourcing trajectory segment.

[0078] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0079] transforming the trajectory direction of each single trajectory segment in the crowdsourced trajectory segment to the main direction, and determining the main direction of the road segment where the crowdsourced trajectory segment is located;

[0080] The step of generating corresponding elevation information according to the height dimension information and the main direction comprises:

[0081] Obtaining height dimension information of the road section where the crowdsourced trajectory segment is located;

[0082] Denoising and fusing the main direction of the road section where the crowdsourced trajectory segment is located and the height dimension information of the road section where the crowdsourced trajectory segment is located are performed to determine the elevation information of the road section where the crowdsourced trajectory segment is located.

[0083] Furthermore, when the elevation information generation program stored in the memory 1005 is executed by the processor, the following steps are also implemented:

[0084] Matching the ground information of the crowd-sourced trajectory to determine the plane map corresponding to the crowd-sourced trajectory;

[0085] If the crowdsourced track is located in the intersection area of ​​the planar map, obtaining the intersection adjacent section corresponding to the crowdsourced track in the planar map;

[0086] Acquire the height information of the adjacent road sections of the intersection, perform floor height calculation and trajectory stretching on the height information of the adjacent road sections of the intersection, and determine the height dimension information corresponding to the intersection area in the plane map.

[0087] To better understand the above technical solutions, exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. Instead, these embodiments are provided to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art.

[0088] Based on the above terminal device architecture but not limited to the above architecture, a specific embodiment of the elevation information generation method of the present invention is proposed.

[0089] Specifically, refer to Figure 2 , Figure 2 1 is a flow chart of a first embodiment of a method for generating elevation information according to the present invention. The method for generating elevation information includes:

[0090] Step S10, obtaining a crowdsourcing trajectory and extracting feature point information of the crowdsourcing trajectory;

[0091] It should be noted that, in this embodiment, the height feature points in the single track are selected according to the height change information of the single track data of each single track in the crowdsourced track, and the feature point information of the height feature points in the single track is extracted according to the height feature points. After obtaining the height feature points of each single track and the feature point information corresponding to the height feature points, the feature point information of the crowdsourced track is determined.

[0092] Optionally, the corresponding feature point information is determined based on the trajectory data of the crowdsourced trajectory by obtaining the height information in the single trajectory data corresponding to each single trajectory from the trajectory data of the crowdsourced trajectory, and obtaining the height change information of the single trajectory based on the height information in the single trajectory data, and determining the height feature points of the single trajectory based on the height change nodes in the height change information, and determining the height feature points in the crowdsourced trajectory by performing data fusion on the height feature points in each single trajectory, and obtaining the corresponding feature point information.

[0093] Specifically, for example, a road section from point A to point B, where point A is a point on the road outside the parking lot where vehicles travel, and point B is a point inside the parking lot where vehicles stop. There are several single trajectories of vehicles traveling from point A to point B, which constitute the crowdsourcing trajectory corresponding to point A to point B. By identifying the slope points and intersections in the single trajectory corresponding to point A to point B, and obtaining the feature point information of the corresponding slope points and intersections in the crowdsourcing trajectory from point A to point B.

[0094] Step S20, calculating the layer height according to the feature point information, determining the corresponding inter-layer height value, and stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information;

[0095] It should be noted that in this embodiment, feature point information for multiple individual track data sets within the crowdsourced trajectory is obtained and statistically analyzed to determine the inter-layer heights corresponding to each individual track during its travel. This information is then integrated and calculated to obtain the inter-layer height values ​​corresponding to the crowdsourced trajectory during its travel. Furthermore, a stretch factor for stretching the crowdsourced trajectory is calculated based on the inter-layer height values ​​corresponding to the crowdsourced trajectory during its travel. By applying the stretch factor to the crowdsourced trajectory and performing an inter-layer height-dimensional stretching transformation, the height-dimensional information of the crowdsourced trajectory is determined.

[0096] Specifically, for example, there is a certain height feature point E in each single trajectory from point C to point D, and its corresponding feature point information includes at least an inter-layer height value. Before the height feature point E, point C is located at the negative first floor and point D is located at the negative second floor. However, for each single trajectory from point C to point D in the crowdsourcing trajectory, there is different inter-layer height information. Point D located at the negative second floor has D1 and D2 with different inter-layer height values. The stretching factor is determined by calculating according to the inter-layer height values ​​of D1 and D2, and each single trajectory in the crowdsourcing trajectory is stretched based on the stretching factor to determine the height dimension information during the movement of the crowdsourcing trajectory.

[0097] Step S30, splitting the crowdsourced trajectory into corresponding planar maps, and obtaining crowdsourced trajectory segments corresponding to each road section of the planar map;

[0098] It should be noted that, in this embodiment, by splitting the crowdsourcing trajectory into the various road sections corresponding to the plane map, the crowdsourcing trajectory segments corresponding to each road section in the plane map are determined, and the height dimension information corresponding to the crowdsourcing trajectory is split into the corresponding crowdsourcing trajectory segments.

[0099] Step S40 , performing principal component analysis (PCA) on the crowd-sourced trajectory segments to determine the main directions of the crowd-sourced trajectory segments, and generating corresponding elevation information based on the height dimension information and the main directions.

[0100] It should be noted that in this embodiment, the PCA analysis is performed on each single trajectory segment in the crowdsourced trajectory segmentation to determine the trajectory direction of each single trajectory. After determining the trajectory direction of each single trajectory in the crowdsourced trajectory, the PCA analysis is performed on the trajectory directions of all single trajectories to determine the main direction of the road segment corresponding to the crowdsourced trajectory segment based on the crowdsourced trajectory in the planar map.

[0101] Furthermore, the height dimension information and main direction of the crowdsourced trajectory segments are denoised and fused to determine the elevation information of the road section in the plane map. The elevation information of the entire plane map is determined based on the height dimension information and main direction of each crowdsourced trajectory segment in the crowdsourced trajectory corresponding to the plane map.

[0102] This embodiment uses a crowdsourced trajectory-based method to determine elevation information, reducing reliance on high-precision detection equipment and lowering equipment costs during data collection. By calculating and height-dimensionally stretching the data of each single track of the crowdsourced trajectory, unified elevation information is obtained, solving the problem of inconsistent quality of collected data, reducing the difficulty of elevation information collection and calculation, and reducing data deviations during the calculation and inference process, thereby improving the accuracy of elevation information.

[0103] Furthermore, based on the first embodiment of the elevation information generating method of the present invention, a second embodiment of the elevation information generating method of the present invention is proposed.

[0104] In this embodiment, the feature point information of the crowdsourcing trajectory is extracted in step S10, and the Figure 3 , specifically including:

[0105] Step S11, obtaining single track data corresponding to the single track according to the crowdsourced track;

[0106] Step S12: determining height feature points according to the height change information of the single-track data, and extracting feature point information of the height feature points.

[0107] It should be noted that in this embodiment, by obtaining the single track data corresponding to each single track in the crowdsourced track, the height change information of the single track calculated based on the single track floor height information in the single track data is determined, and the height feature points in the single track are determined based on the height change information of the single track, where the height feature points refer to the change nodes existing in the height change information of the single track. After determining the height feature points in the single track, the feature point information of the height feature points is obtained, where the feature point information includes at least the inter-layer height of the height feature points in the single track, the coordinates of the height feature points in the plane map, etc.

[0108] Furthermore, in step S10, after extracting the characteristic point information of the crowdsourced trajectory, the elevation information generation method further includes obtaining the height information of adjacent sections of each intersection in the intersection area to calculate the floor height and stretch the trajectory if the crowdsourced trajectory is located in the intersection area, referring to Figure 4 , specifically including:

[0109] Step S101, matching the ground information of the crowdsourcing trajectory to determine the plane map corresponding to the crowdsourcing trajectory;

[0110] It should be noted that, in this embodiment, by matching the crowdsourcing trajectory with the plane map, the crowdsourcing trajectory of each road section in the plane map and the road section corresponding to the crowdsourcing trajectory in the plane map are determined, so as to achieve matching and alignment of the crowdsourcing trajectory with the plane map. Specifically, the way to match the crowdsourcing trajectory with the plane map can be to match the ground road sign information such as parking spaces, ground signs, lane lines, speed bumps, etc. mounted on the crowdsourcing trajectory with the 2D map, and to achieve alignment of the trajectory data of the crowdsourcing trajectory with the 2D map based on the road feature information mounted in the crowdsourcing trajectory.

[0111] Step S102: if the crowdsourced track is located in the intersection area of ​​the planar map, obtaining the intersection adjacent section corresponding to the crowdsourced track in the planar map;

[0112] Step S103 , obtaining height information of adjacent road sections of the intersection, performing floor height calculation and trajectory stretching on the height information of adjacent road sections of the intersection, and determining height dimension information corresponding to the intersection area in the plane map.

[0113] In this embodiment, by aligning the crowdsourcing trajectory with the corresponding plane map, the corresponding crowdsourcing trajectory segments in each road section in the plane map are determined. When the crowdsourcing trajectory is located in the intersection area of ​​the plane map (the intersection area refers to the intersection area connected by n roads or road sections, n>2), the height dimension information of the intersection area in the plane map is calculated according to the pre-set intersection area height dimension information.

[0114] Optionally, the height dimension information of the intersection area in the plane map is calculated based on the pre-set intersection area height dimension information by obtaining the adjacent road sections of the road section where the crowdsourcing trajectory is located, and performing height dimension calculation based on the trajectory data corresponding to the crowdsourcing trajectory of the adjacent road sections of the road section in the intersection area where the crowdsourcing trajectory is located, to determine the height dimension information of the intersection area in the plane map.

[0115] Specifically, the height dimension calculation method for the trajectory data corresponding to the adjacent road section can be to obtain the layer height information of each single trajectory corresponding to the adjacent road section, perform linear interpolation trajectory stretching according to the layer height information of each single trajectory corresponding to the adjacent road section, obtain the height information of the adjacent road section, and determine the height dimension information of the intersection area where the crowdsourced trajectory is located.

[0116] This embodiment achieves matching of crowdsourced trajectories with plane maps by extracting height feature points from crowdsourced trajectories and obtaining feature point information of the height feature points. It extracts height feature points through the trajectory data of each single trajectory, and obtains the corresponding feature point information for subsequent height dimension information calculation, thereby reducing the difficulty of height dimension information collection, calculation, and unification. It performs preset height dimension calculation on the crowdsourced trajectories in the intersection area, innovates the calculation method of height dimension information in the intersection area, optimizes the alignment effect of height alignment in the intersection area, and improves the accuracy of height dimension information in the intersection area.

[0117] Furthermore, based on the first and second embodiments of the elevation information generating method of the present invention, a third embodiment of the elevation information generating method of the present invention is proposed.

[0118] In this embodiment, step S20 is performed to calculate the layer height according to the feature point information, determine the corresponding inter-layer height value, and stretch the crowdsourced trajectory according to the inter-layer height value to determine a refinement scheme for the height dimension information.

[0119] Reference Figure 5 In step S20, the floor height is calculated based on the feature point information to determine the corresponding inter-floor height value, specifically including:

[0120] Step S201, obtaining feature point information of several single trajectories of the crowdsourced trajectory;

[0121] It should be noted that, in this embodiment, the crowdsourced trajectory is trajectory information containing several single trajectories. By obtaining the feature point information for a certain height feature point in each single trajectory, extracting the floor height information from the feature point information of each single trajectory, and confirming the floor height value corresponding to each single trajectory.

[0122] Step S202: performing statistics on the feature point information corresponding to the plurality of single tracks to determine the inter-layer heights of the plurality of single tracks;

[0123] In this embodiment, the floor height value corresponding to each single track is counted. During the statistical process, the floor height value of each single track is arranged in order. The floor height value of the single track that is arranged in order refers to counting the height value of a certain height feature point of each single track in the crowdsourced track, and realizing the orderly sorting of the height value of a certain height feature point of each single track.

[0124] Step S203 , performing layer height calculation on the inter-layer heights of the plurality of single trajectories to determine an inter-layer height value of the crowdsourced trajectory.

[0125] It should be noted that the inter-layer height value of each single track is averaged by using the preset inter-layer height calculation rule to determine the inter-layer height obtained by calculating several single tracks, and this inter-layer height value is used for the inter-layer height value calculated by the subsequent crowdsourcing track.

[0126] Optionally, the mean value of the floor height of each single track can be calculated by obtaining the floor height value of each single track and obtaining a floor height value sequence [h1,h2,h3,...h n ], calculate the average value u and variance v of the storey height according to all the storey height values ​​in the storey height value sequence, determine the selection range of the storey height value [uv,u+v] by the average value u and variance v of the storey height, and select the storey height value in the storey height value sequence [h1,h2,h3,...h n ] that fall within the selected range [uv,u+v], and calculate the average of all the layer height values ​​that fall within the selected range to determine the final layer height average value m, and use the layer height average value m as the inter-layer height value calculated by the crowdsourcing trajectory.

[0127] Further, refer to Figure 6 In step S20, the crowdsourcing trajectory is stretched according to the inter-layer height value to determine the height dimension information, which specifically includes:

[0128] Step S204: Substitute the interlayer height value into a preset formula to calculate and determine the interlayer stretching factor;

[0129] It should be noted that the above execution step uses the calculated average floor height m as the inter-floor height value for calculating the crowdsourcing trajectory. In this embodiment, the inter-floor stretching factor of the single trajectory is obtained by calculating the inter-floor height value (average floor height m) and the height value of the single trajectory in the crowdsourcing trajectory.

[0130] Optionally, the inter-layer height value corresponding to the crowdsourcing track where the single track 1 is located and the inter-layer value of the single track 1 are substituted into the preset formula of the inter-layer stretching factor to calculate and determine the inter-layer stretching factor for the track stretching calculation performed by the single track 1 on the crowdsourcing track. Specifically, for example, in an exemplary embodiment, the inter-layer height value of the single track 1 and the inter-layer height value corresponding to the crowdsourcing track where the single track 1 is located are [h1,h2] respectively. m ], then according to the layer height value h1 of the single track 1 and the layer height value h of the crowdsourcing track m Calculate and determine the corresponding interlayer stretching factor α = h m / h1.

[0131] Step S205, performing height dimension stretching calculation on the single track data corresponding to the single track according to the inter-layer stretching factor to determine the height dimension information of the single track;

[0132] It should be noted that, in this embodiment, the height dimension information of the single track after track stretching is obtained by substituting the layer height value of each single track in the crowdsourced track and the inter-layer height value corresponding to each single track into the preset height stretching formula for calculation. Specifically, for example, in an exemplary embodiment, the height dimension information of the single track after track stretching is obtained by substituting the initial height information z0 of the single track 1, the layer height value z of the single track after the height feature point, and the inter-layer stretching factor α into the corresponding calculation to determine the height dimension information new_z=z0+α-(z-z0)α of the single track.

[0133] Specifically, for example, referring to Figure 7 , Figure 7 This is an exemplary specific flow chart of this embodiment, where z0, z1, and z2 represent specific height values ​​on the height axis z, h1 represents the floor height value of a single track, and h m is the inter-layer height value of the crowdsourcing trajectory where the single trajectory is located.

[0134] Step S206 : determining the height dimension information of the crowdsourcing trajectory based on the height dimension information of the single trajectory.

[0135] It should be noted that, in this embodiment, the height dimensional information corresponding to the crowdsourcing trajectory is determined by performing linear interpolation calculation on the height dimensional information of each single trajectory in the crowdsourcing trajectory. Specifically, the linear interpolation method for calculation can be to perform statistics on the height dimensional information of each single trajectory, calculate the corresponding mean and variance, determine the selection range of the crowdsourcing trajectory height dimensional information, and calculate the average value of the height dimensional information within the selection range, and use the average value as the height dimensional information of the crowdsourcing trajectory.

[0136] This embodiment calculates and stretches the height dimension of each single track data of the crowdsourced track to obtain unified height dimension information. The crowdsourced data supplementation method expands the data source channels, realizes the mutual non-aggregation and mutual verification of different data, and obtains more accurate height information based on the crowdsourced data. This reduces the difficulty of collecting, calculating, and unifying height dimension information, improves the accuracy of height dimension information, and reduces data deviation in the calculation and inference process.

[0137] Furthermore, based on the first, second and third embodiments of the elevation information generating method of the present invention, a fourth embodiment of the elevation information generating method of the present invention is proposed.

[0138] This embodiment performs principal component analysis (PCA) on the crowdsourced trajectory segments in step S40 to determine the main direction of the crowdsourced trajectory segments, and generates corresponding elevation information refinement based on the height dimension information and the main direction. Figure 8 , specifically including:

[0139] Step S41, performing a PCA analysis on each single trajectory segment in the crowdsourced trajectory segmentation to determine the trajectory direction corresponding to each single trajectory segment;

[0140] It should be noted that in this embodiment, PCA analysis is performed on each road section in the planar map based on the crowdsourced trajectory segmentation. There are several single trajectory data in the crowdsourced trajectory segment corresponding to each road section. In this embodiment, PCA first direction analysis is first performed on each single trajectory in the crowdsourced trajectory segment to determine the trajectory direction of each single trajectory in the crowdsourced trajectory segment.

[0141] Specifically, PCA (principal components analysis) aims to use the idea of ​​dimensionality reduction to transform multiple indicators into a few comprehensive indicators. PCA analysis is a technology for simplifying data sets. It switches the data into a new coordinate system through linear transformation, so that the first largest variance of any data projection is on the first coordinate (called the first principal component), the second largest variance is on the second coordinate (the second principal component), and so on. Principal component analysis can reduce the dimensionality of the data set while maintaining the features that contribute the most to the variance of the data set, and can retain the most important aspects of the data. This embodiment retains the main trajectory direction in a single trajectory through PCA analysis.

[0142] Step S42 : performing a PCA secondary analysis on the trajectory direction corresponding to each single trajectory segment in the crowdsourcing trajectory to determine the main direction of the crowdsourcing trajectory segment.

[0143] The first direction analysis analyzes a single trajectory, and the second direction analysis analyzes the trajectory directions of all single trajectories determined based on the first direction analysis.

[0144] It should be noted that, in this embodiment, after the PCA first direction analysis is performed on each single track in the crowdsourced trajectory segment, each single track in the crowdsourced trajectory segment has its corresponding trajectory direction. However, the road section in the planar map where the single track is located also includes several other single tracks and the trajectory directions corresponding to several other single tracks. In this embodiment, the trajectory directions corresponding to the several single tracks are analyzed in the second direction by PCA, and the main direction of the crowdsourced trajectory segment is determined according to the trajectory direction of each single track.

[0145] Specifically, for example, referring to Figure 9 For the crowdsourcing trajectory segment, there are four single trajectories t1, t2, t3, and t4. PCA first direction analysis is performed on these four single trajectories to determine the trajectory directions p1, p2, p3, and p4 corresponding to the four single trajectories t1, t2, t3, and t4. Then, PCA second direction analysis is performed on the trajectory directions p1, p2, p3, and p4 corresponding to the four single trajectories to determine the main direction P corresponding to the crowdsourcing trajectory segment.

[0146] Furthermore, in step S42, after PCA second direction analysis is performed on the trajectory directions of all single trajectory segments in the crowdsourced trajectory to determine the main direction of the crowdsourced trajectory segment, the method further includes performing denoising and fusion based on the main direction and height dimension information of the crowdsourced trajectory to obtain the elevation information of the crowdsourced trajectory, which specifically includes:

[0147] Step S43, transforming the trajectory direction of each single trajectory segment in the crowdsourced trajectory segment to the main direction, and determining the main direction of the road segment where the crowdsourced trajectory segment is located;

[0148] It should be noted that, in this embodiment, after determining the main direction of the crowdsourced trajectory segment of the road section where the crowdsourced trajectory segment is located, all single trajectories of the crowdsourced trajectory segment are respectively transformed to the main direction of the crowdsourced trajectory, and the road section corresponding to the crowdsourced trajectory segment in the plane map is determined. Specifically, for example, the trajectory directions of the single trajectories t1, t2, t3, and t4 are transformed so that the transformed p1, p2, p3, and p4 are consistent with the main direction P of the crowdsourced trajectory, and the main direction of the road section where the crowdsourced trajectory segment is located in the plane map is determined.

[0149] Step S44, obtaining the height dimension information of the road section where the crowdsourced trajectory segment is located;

[0150] In this embodiment, after determining the main direction of the road section where the crowdsourced trajectory segment is located, the height dimension information corresponding to the crowdsourced trajectory segment is extracted, and based on the height dimension information and the main direction of the crowdsourced trajectory segment, the elevation information of the road section corresponding to the crowdsourced trajectory segment in the plane map is determined.

[0151] Step S45 , performing denoising and fusion on the main direction of the road section where the crowdsourced trajectory segment is located and the height dimension information of the road section where the crowdsourced trajectory segment is located, to determine the elevation information of the road section where the crowdsourced trajectory segment is located.

[0152] It should be noted that, in this embodiment, the elevation information of the road section corresponding to the crowdsourced trajectory segment is formed by denoising and fusing the height dimension information and the main direction of the crowdsourced trajectory segment.

[0153] Optionally, this embodiment adopts the NURBS spline curve fitting method to perform smooth filtering on the height dimension information, thereby achieving smooth filtering of the height dimension information in all directions in the plane map. In addition, the height dimension information can also be used as one-dimensional information, and smooth filtering of the height dimension information can be achieved by adopting filtering schemes such as mean filtering and Gaussian filtering. The height change information of the crowdsourced trajectory segment in the main direction of the road section in the plane map is determined, and accurate height dimension information is written to each road section in the plane map through the height dimension information and the corresponding main direction. The section information of each road section is integrated for fusion mapping to construct a three-dimensional map with direction and corresponding height dimension information.

[0154] This embodiment performs a secondary PCA directional analysis on the crowdsourced trajectories to obtain the main direction of each road section in the planar map, and then performs denoising and fusion based on the main direction and height dimension information of each road section to automatically generate the corresponding elevation information. This reduces data deviation in the calculation and inference process and improves the precision and accuracy of the elevation information.

[0155] In addition, the embodiment of the present invention also proposes an elevation information generating device, referring to Figure 10 , Figure 10 Schematic diagram of the functional modules of the elevation information generating device of the present invention, the elevation information generating device includes:

[0156] An acquisition module 10 is used to acquire a crowdsourcing trajectory and extract feature point information of the crowdsourcing trajectory;

[0157] A calculation module 20 is configured to calculate the floor height according to the feature point information, determine the corresponding inter-floor height value, and stretch the crowdsourced trajectory according to the inter-floor height value to determine the height dimension information;

[0158] A splitting module 30 is used to split the crowdsourced trajectory into corresponding planar maps and obtain crowdsourced trajectory segments corresponding to each road section of the planar map;

[0159] The analysis module 40 is configured to perform principal component analysis (PCA) on the crowdsourced trajectory segments to determine the main directions of the crowdsourced trajectory segments, and generate corresponding elevation information based on the height dimension information and the main directions.

[0160] In addition, an embodiment of the present invention also proposes a device, which includes a memory, a processor, and an elevation information generation program stored in the memory and runnable on the processor. When the elevation information generation program is executed by the processor, the steps of the elevation information generation method described above are implemented.

[0161] The various embodiments of the elevation information generating device and the computer-readable storage medium of the present invention may refer to the various embodiments of the elevation information generating method of the present invention, and will not be described in detail here.

[0162] The specific embodiments of the computer program product of the present invention are basically the same as the embodiments of the above-mentioned elevation information generating method, and will not be described in detail here.

[0163] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or apparatus comprising the element.

[0164] The serial numbers of the above embodiments of the present invention are for description only and do not represent the advantages or disadvantages of the embodiments.

[0165] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better embodiment. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a device (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in each embodiment of the present invention.

[0166] The above are only preferred embodiments of the present invention and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention description and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for generating elevation information, characterized in that: The elevation information generation method comprises: Obtaining a crowdsourcing trajectory and extracting feature point information of the crowdsourcing trajectory; Calculating the layer height according to the feature point information to determine the corresponding inter-layer height value, and stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information; Splitting the crowdsourced trajectory into corresponding planar maps, and obtaining crowdsourced trajectory segments corresponding to each road section of the planar map; A principal component analysis (PCA) is performed on the crowdsourced trajectory segments to determine the main directions of the crowdsourced trajectory segments, and corresponding elevation information is generated according to the height dimension information and the main directions.

2. The method for generating elevation information according to claim 1, wherein: The step of extracting feature point information of the crowdsourcing trajectory includes: According to the crowdsourced trajectory, obtaining single trajectory data corresponding to the single trajectory; According to the height change information of the single track data, a height feature point is determined, and feature point information of the height feature point is extracted.

3. The method for generating elevation information according to claim 2, wherein: The step of calculating the layer height according to the feature point information and determining the corresponding layer height value includes: Obtaining feature point information of several single trajectories of the crowdsourced trajectory; Performing statistics on the feature point information corresponding to the plurality of single tracks to determine the inter-layer heights of the plurality of single tracks; The inter-layer heights of the plurality of single trajectories are calculated to determine an inter-layer height value of the crowdsourced trajectory.

4. The method for generating elevation information according to claim 3, wherein: The step of stretching the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information includes: Substituting the interlayer height value into a preset formula for calculation to determine the interlayer stretching factor; performing a height dimension stretching calculation on the single track data corresponding to the single track according to the inter-layer stretching factor to determine the height dimension information of the single track; Based on the height dimension information of the single trajectory, the height dimension information of the crowdsourced trajectory is determined.

5. The method for generating elevation information according to claim 1, wherein: The step of performing principal component analysis (PCA) on the crowdsourcing trajectory segments to determine the main directions of the crowdsourcing trajectory segments includes: Performing a PCA analysis on each single trajectory segment in the crowdsourced trajectory segmentation to determine the trajectory direction corresponding to each single trajectory segment; A PCA secondary analysis is performed on the trajectory direction corresponding to each single trajectory segment in the crowdsourcing trajectory to determine the main direction of the crowdsourcing trajectory segment.

6. The method for generating elevation information according to claim 5, wherein: After the step of determining the main direction of the road segment where the crowdsourced trajectory segment is located, the method further includes: transforming the trajectory direction of each single trajectory segment in the crowdsourced trajectory segment to the main direction, and determining the main direction of the road segment where the crowdsourced trajectory segment is located; The step of generating corresponding elevation information according to the height dimension information and the main direction comprises: Obtaining height dimension information of the road section where the crowdsourced trajectory segment is located; Denoising and fusing the main direction of the road section where the crowdsourced trajectory segment is located and the height dimension information of the road section where the crowdsourced trajectory segment is located are performed to determine the elevation information of the road section where the crowdsourced trajectory segment is located.

7. The method for generating elevation information according to claim 1, wherein: After the step of obtaining the crowdsourcing trajectory, the method further includes: Matching the ground information of the crowd-sourced trajectory to determine the plane map corresponding to the crowd-sourced trajectory; If the crowdsourced track is located in the intersection area of ​​the planar map, obtaining the intersection adjacent section corresponding to the crowdsourced track in the planar map; Acquire the height information of the adjacent road sections of the intersection, perform floor height calculation and trajectory stretching on the height information of the adjacent road sections of the intersection, and determine the height dimension information corresponding to the intersection area in the plane map.

8. An elevation information generating device, characterized in that: The elevation information generating device comprises: An acquisition module, configured to acquire a crowdsourcing trajectory and extract feature point information of the crowdsourcing trajectory; a calculation module, configured to calculate the layer height according to the feature point information, determine the corresponding inter-layer height value, and stretch the crowdsourced trajectory according to the inter-layer height value to determine the height dimension information; A splitting module is used to split the crowdsourced trajectory into corresponding plane maps and obtain crowdsourced trajectory segments corresponding to each road section of the plane map; The analysis module is used to perform PCA direction analysis on the crowd-sourced trajectory segment, determine the main direction of the crowd-sourced trajectory segment, and generate corresponding elevation information based on the height dimension information and the main direction.

9. A device, characterized in that The device includes a memory, a processor, and an elevation information generation program stored in the memory and executable on the processor. When the elevation information generation program is executed by the processor, the steps of the elevation information generation method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores an elevation information generation program, which, when executed by a processor, implements the steps of the elevation information generation method according to any one of claims 1 to 7.

Images