CN110379007B - Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud - Google Patents
Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud Download PDFInfo
- Publication number
- CN110379007B CN110379007B CN201910674347.7A CN201910674347A CN110379007B CN 110379007 B CN110379007 B CN 110379007B CN 201910674347 A CN201910674347 A CN 201910674347A CN 110379007 B CN110379007 B CN 110379007B
- Authority
- CN
- China
- Prior art keywords
- road
- curve
- vehicle
- point
- laser scanning
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 57
- 238000013461 design Methods 0.000 claims abstract description 18
- 238000000605 extraction Methods 0.000 claims abstract description 14
- 238000010845 search algorithm Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 3
- 241000245165 Rhododendron ponticum Species 0.000 claims 1
- 238000010276 construction Methods 0.000 abstract description 7
- 238000013441 quality evaluation Methods 0.000 abstract description 5
- 238000013507 mapping Methods 0.000 description 6
- 239000013598 vector Substances 0.000 description 5
- 238000013178 mathematical model Methods 0.000 description 4
- 206010039203 Road traffic accident Diseases 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 241000208422 Rhododendron Species 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000001303 quality assessment method Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 241000544061 Cuculus canorus Species 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud, which comprises the following steps: s1, dividing the vehicle-mounted mobile laser scanning original point cloud into a plurality of segments of sub-point clouds; s2, performing pavement extraction on each segment of sub-point cloud to obtain a pavement point set; s3, extracting road signs from the road point set to obtain a road sign point set; s4, clustering the road sign point set to obtain a plurality of road sign sub-point sets; s5, performing shell extraction on each road sign sub-point set to obtain a shell point set; s6, respectively carrying out model fitting on a basic three-dimensional road curve model and a shell point set specified by a standard to obtain a candidate curve; and S7, comparing the candidate curves to obtain an optimal curve. The method can automatically and robustly identify the type of the road curve and accurately estimate the parameters of the three-dimensional road curve meeting the geometric design standard of the road under the condition that the data contains a large number of outliers, and can be used in various fields such as road construction quality evaluation.
Description
Technical Field
The invention relates to a three-dimensional road curve reconstruction method, in particular to a method for reconstructing a three-dimensional road curve by utilizing vehicle-mounted mobile laser scanning point cloud.
Background
The curved shape of the road has an important influence on road construction, traffic safety and driving comfort. The goal of road curve reconstruction is to estimate the geometric information (i.e., the type of curve and parameter values) of the curve of the constructed road. Therefore, road curve reconstruction can be applied to digital mapping, automatic driving, road geometric design, traffic accident analysis, road construction quality assessment and the like.
There are two main types of data currently available for reconstructing highway curves, the first being ground survey data, mainly using real-time kinematic GPS (global positioning system) technology; the second is to measure data off-ground using digital images or the like acquired by remote sensors installed on an airplane or a satellite. Non-terrestrial survey data can cover a larger geographic area at a lower cost than terrestrial survey data, but its spatial accuracy is typically lower than terrestrial survey data.
Mobile mapping vehicles with on-board optical cameras and integrated GPS and IMU (inertial measurement unit) systems are becoming better solutions providing higher accuracy and higher efficiency. The trajectory data of the mobile mapping vehicle and the collected images may be used to detect and reconstruct the highway centerline. However, these highway centerlines have only limited application. For example, with the development of recent automatic driving, the automatic driving automobile needs not only a center line of a road, but also a side line of the road, a lane line, a road mark and the like, which are key elements of a future high-definition map. In recent years, a Mobile Laser Scanning (MLS) system can rapidly acquire a three-dimensional point cloud with and rich road surface information. Therefore, compared with a camera, the vehicle-mounted mobile laser scanning system has great advantages in road curve reconstruction.
There are some methods proposed to reconstruct road geometry using vehicle-mounted mobile laser scan data, such as: estimating the superelevation by using rectangular areas among the road surface marks; dividing the vehicle-mounted mobile laser scanning data into a plurality of blocks by using the trajectory and the scanning angle at the same time, and then estimating the slope and the superelevation of each block by using a principal component analysis algorithm; thirdly, extracting road marks by using the intensity information and the scanning angle, and then calculating an azimuth angle and a curvature map by using the extracted mark points to estimate a horizontal curve variable; and fourthly, dividing the vehicle-mounted mobile laser scanning data into a plurality of layer blocks by using the track data so as to estimate the slope and the superelevation. In the methods, horizontal or vertical (two-dimensional) curves are respectively reconstructed from vehicle-mounted mobile laser scanning data, but roads are three-dimensional objects, and the curves are three-dimensional in nature, and meanwhile, a lot of applications currently need comprehensive three-dimensional road geometric information (such as road geometric design, traffic accident analysis, highway quality evaluation and the like). In addition, many methods reconstruct curves that are general curves (e.g., basic spline curves), so that these methods cannot be used for road construction quality assessment. Since the road geometry design criteria only specify that certain curves can be used to design and build roads. Current road geometric design standards specify five basic types of horizontal road curves (i.e., straight, left, right, left-handed and right-handed) and three basic types of vertical road curves (i.e., straight, concave and convex parabolas). The combination of the substantially horizontal curve and the substantially vertical curve forms a substantially three-dimensional road curve.
In summary, there is a need for a method for automatically reconstructing a three-dimensional curve meeting the geometric design standard of a road from a vehicle-mounted mobile laser scanning point cloud.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a three-dimensional road curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud.
The invention specifically adopts the following technical scheme:
a three-dimensional road curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud comprises the following steps:
s1, dividing the vehicle-mounted mobile laser scanning original point cloud into a plurality of sub-point clouds by using the driving track information of the laser scanning vehicle;
s2, carrying out road surface extraction on each section of sub-point cloud obtained in the step S1 to obtain a road surface point set;
s3, extracting road signs from the road surface point set obtained in the step S2 to obtain a road sign point set;
s4, clustering the road sign point sets obtained in the step S3 to obtain a plurality of road sign sub-point sets;
s5, performing shell extraction on each road sign sub-point set obtained in the step S4 to obtain a shell point set;
s6, performing model fitting on 15 basic three-dimensional road curve models specified by the road geometric design standard and the shell point set obtained in the step S5 to obtain 15 candidate curves;
and S7, comparing the candidate curves obtained in the step S6 to obtain an optimal curve.
Further, the specific process of step S1 is; inputting a vehicle-mounted mobile laser scanning original point cloud G and a track point cloud T corresponding to the original point cloud G, firstly dividing the T into J subsets according to the storage sequence of points: t is1,T2,…,TJAfter segmentation, except for the last subset TJEach of the remaining subsets has E points, and then G is divided into J segments G as follows1,G2,…,GJWherein
Where d (u, v) represents the Euclidean distance between two three-dimensional points, R ∈ R is the distance threshold used to determine the neighborhood of points, G1And T1Corresponding 1 st set of points, G, divided from GjAnd TjCorresponding j-th set of points, T, divided from GjIs the jth subset divided from Tj-1Is the j-1 th subset divided from T.
Further, the distance threshold r for determining the neighborhood of points is 10 meters.
Further, the specific process of step S2 is: setting a smoothness threshold, extracting the road surface by using a smoothness-based region growing algorithm, obtaining a plurality of sub-point sets with different sizes after region growing, and considering the sub-point set with the largest number of points as the extracted road surface point set.
Still further, the smoothness threshold is set to 2 degrees.
Further, the implementation process of step S3 includes: an intensity variance threshold is set and an Otsu threshold algorithm based on intensity variance is used to extract road signs.
Further, in step S4, the road sign point set extracted in step S3 is clustered by using euclidean clustering to obtain a plurality of road sign sub-point sets.
Further, in step S5, a shell point set is extracted from the road sign sub-point set obtained in step S4 by an alpha shape (α -shape) method.
Further, the specific implementation process of step S6 is as follows: model fitting is respectively carried out on 15 basic three-dimensional road curve models specified by road geometric design standards and the shell point set obtained in the step S5 to obtain 15 candidate curves, and for each curve model m ∈ {1,2, …,15}, the model fitting method tries to find the following candidate curves:
wherein C ismIs the set of curves defined by the model m and D is the set of data points. f (C, D) is the objective function used for the fitting, i.e. the mean measure:
f(C,D)＝g(C)/d2(C,D)
wherein g (C) represents the arc length of curve C, d2(C, D) is the one-way modified Hausdorff distance from C to D:
wherein C issIs a set of points uniformly sampled from C with very little resolution, | CSThe number of points in the point set is represented by | p-q | represents the modulus of the vector, and the optimization problem (i.e., the model fitting problem) shown above is solved by a rhododendron search algorithm.
Further, the optimum curve obtained in step S7 is
Wherein
After adopting the technical scheme, compared with the background technology, the invention has the following advantages:
1. the method can automatically reconstruct a three-dimensional road curve from the vehicle-mounted mobile laser scanning point cloud, and the reconstructed curve is a curve specified by a road geometric design standard, so that the method is not only suitable for digital mapping, but also suitable for road construction quality evaluation; secondly, the road curve reconstruction is realized by using vehicle-mounted mobile laser scanning data, and richer and more reliable three-dimensional information about the road can be obtained; finally, the present invention directly reconstructs three-dimensional road curves, while most of the existing methods reconstruct two-dimensional road curves separately, and compared to two-dimensional methods, the present invention can be easily extended to deal with these new three-dimensional road curves.
2. In terms of data preprocessing, the present invention proposes to extract road markings using only the variance of intensity as a feature. Therefore, the proposed road sign extraction method is simpler and more convenient than the existing extraction method.
3. In the aspect of road curve reconstruction, the invention adopts a robust probabilistic program inference technology, thereby being capable of automatically and robustly identifying the type of the curve and accurately estimating the parameters of the curve under the condition of containing outliers.
Drawings
FIG. 1 is a schematic flow chart of an embodiment of the present invention;
FIG. 2 illustrates raw point cloud data according to an embodiment of the present invention;
fig. 3 is a three-dimensional road curve reconstructed according to the embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Examples
Fig. 1 shows an implementation process of this embodiment, and referring to fig. 1, the invention discloses a three-dimensional road curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud, which includes the following steps:
s1, dividing the vehicle-mounted mobile laser scanning original point cloud into a plurality of sections of sub-point clouds by using the driving track information of the laser scanning vehicle:
this step, i.e., chunking the data, as shown in fig. 2, typically involves a large number (millions or tens of millions) of points in the original vehicle-mounted moving laser scanning point cloud, which is difficult to process at the same time. The common practice is to divide the original vehicle-mounted mobile laser scanning point cloud into a plurality of small point clouds for processing respectively. The goal of this embodiment is road curve reconstruction, so vehicle-mounted mobile laser scanning point cloud partitioning along the road direction is best. As the storage sequence of the points in the point cloud of the vehicle-mounted mobile laser scanning is irregular, the track points (white solid lines adjacent to the middle dotted line of the road in fig. 2) which are sequenced according to the acquisition time can be used for realizing the blocking along the road. The method comprises the steps of enabling an original vehicle-mounted mobile laser scanning point cloud to be G and a corresponding track point set to be T, enabling each point in the G to have four attributes, recording space position information by the three attributes, and recording laser reflection intensity information by the other attribute. Each point in T has three spatial location attributes. Firstly, dividing T into J subsets according to the storage order of points: t is1,T2,…,TJAfter segmentation, except for the last subset TJEach of the remaining subsets has E points, and then G is divided into J segments G as follows1,G2,…,GJ。
Given a three-dimensional point u ═ xu,yu,zu) In G, the set of neighboring points consisting of points adjacent to u is:
wherein d (u, v) represents the Euclidean distance between two three-dimensional points:
r ∈ R is the distance threshold used to determine the neighborhood relationship. In step S1 (i.e., data chunking), r
Set to 10 meters. In accordance with the above definition,
first part G divided from G1And other partial point sets DjThe definition is as follows:
G1and T1Corresponding 1 st set of points, G, divided from GjAnd TjCorresponding j-th set of points, T, divided from GjIs the jth subset divided from Tj-1Is the j-1 th subset divided from T.
S2, performing pavement extraction on each segment of sub-point cloud to obtain a pavement point set:
for road curve reconstruction, only road marks on the road surface are useful objects, and the extraction of road points before the extraction of the road marks can reduce the interference caused by other irrelevant objects. Considering that most of the road surface is smooth, the present embodiment extracts the road surface using a region growing algorithm based on smoothness. In the vehicle-mounted moving laser scanning point cloud, the closer an object is to the laser scanner, the greater the density of points reflected from the surface thereof. In general, the road surface is an object closest to the laser scanner. Thus, in the region growing, the road surface generally corresponds to the region having the largest number of points. That is, the road surface extraction is to extract a region having the largest number of points. In the smoothness-based region growing algorithm, smoothness is measured by a normal vector. Therefore, this step requires two parameters: one is a distance threshold to determine the neighborhood to compute normal vectors and the other is a smoothness threshold to perform normal vector comparisons. The smoothness threshold is set to 2 degrees in this embodiment.
S3, extracting road signs from the road point set to obtain a road sign point set;
s4, clustering the road sign point set to obtain a plurality of road sign sub-point sets;
the steps S3 and S4 are used for road sign extraction and clustering, and since the road sign is generally brighter than its surrounding objects, the reflection intensity of the road sign is generally greater than that of its surrounding objects in the vehicle-mounted mobile laser scanning point cloud. However, the reflection intensity of different road signs in the vehicle-mounted moving laser scanning point cloud can vary greatly due to many factors such as the laser incidence angle and distance. Therefore, simply using one intensity threshold is usually not enough to extract all road marking points.
The final goal of the invention is to reconstruct the road curve, which can be achieved with the edge points identified by the road. The present embodiment proposes a simple method to extract the edge points of the road sign, which is based on the following two observations. Comparing with non-edge points, the variance of the intensity of a neighborhood point set of the edge points is usually larger; ② more importantly, the magnitude of these intensity variance values do not differ much for different edges. That is, the road sign edge point may be extracted by a single intensity variance threshold. The intensity variance of point p is calculated according to the following formula:
wherein D is the set of road surface points extracted in the previous step, i is the intensity of the points,
indicating the number of points in the point set.
After the intensity variance of each point exists, a threshold value can be calculated by the Otsu algorithm to respectively separate the edge point from other points, namely, the point with the intensity variance larger than the threshold value is considered as the edge point. Considering that a section of road surface usually contains a plurality of road signs, the extracted edge points usually do not belong to the same road sign, and the embodiment further uses the euclidean clustering algorithm to cluster the extracted edge points into a plurality of classes. To sum up, this step requires two parameters: a distance threshold is used to determine the neighborhood of points to compute the intensity variance, and a distance threshold is used for clustering.
S5, performing shell extraction on each road sign sub-point set to obtain a shell point set:
the road sign extracted by the invention is wider than the real road sign generally, because the invention uses the intensity variance as the characteristic to extract points, the edge points and the points nearby the edge points are extracted generally. Therefore, the possible error of reconstructing the curve by directly using the extracted road identification points is relatively large. In fact, even the real road sign has a certain width, and cannot be directly used to accurately reconstruct the road curve, and the curve directly reconstructed from all the points of the road sign cannot completely and correctly express the real shape of the road sign. Therefore, it is proposed herein to extract hull points of a road curve by an alpha-shape (α -shape) method to reconstruct the road curve. The alpha shape method requires only one parameter (i.e., alpha). In this example, α is set to 0.1.
S6, performing model fitting on 15 basic three-dimensional road curve models and a shell point set specified by a road geometric design standard to obtain 15 candidate curves;
and S7, comparing the candidate curves to obtain an optimal curve.
Step S6 and step S7 are the curve reconstruction process, and the road geometry design criteria specify five basic types of horizontal road curves (see table 1) and three basic types of vertical road curves (see table 2). The combination of the substantially horizontal curve and the substantially vertical curve forms a substantially three-dimensional road curve. Thus, there are fifteen basic types of three-dimensional road curves.
TABLE 1 mathematical model of the highway level curve
TABLE 2 mathematical model of vertical highway curves
| Vertical curve type | Mathematical model |
| Straight line | z(t)＝z0+ts0 |
| Concave parabola | z(t)＝z0+ts0+t2/(2rV0) |
| Convex parabola | z(t)＝z0+ts0-t2/(2rV0) |
For example, the mathematical model of the basic three-dimensional curve formed by combining the horizontal left-handed spiral and the vertical concave parabola is:
the model is represented in parametric form. The parameter t is equal to [0, L ∈]Where L is the arc length of the curve. Plus L, the model has nine variables. The remaining eight variables are as follows: x is the number of0、y0And z0Is a starting position, α0Is the starting angle, s0Is the initial slope, rV0Is the starting vertical radius, β corresponds to the starting horizontal radius, and γ is the rate of curvature change. Similarly, in table 1, γ is the radius of the horizontal circular curve.
The goal of curve reconstruction is to find a curve that best fits a given set of data points. The curve in the road curve reconstruction is defined by one of the fifteen curve models described above. For the purpose of reconstructing a road curve, the present embodiment first fits the fifteen models to the data in sequence to obtain fifteen candidate curves, and then selects a best-fit curve from the candidate curves. It can be seen that model fitting is a key step in curve reconstruction. Since the extracted hull points may contain a large number of outliers, the present embodiment uses an average measure-based fitting method to robustly cope with these outliers. For each curve model m e {1,2, …,15}, the model fitting method will attempt to find the following candidate curves:
wherein C ismIs the set of curves defined by the model m and D is the set of data points. f (C, D) is the objective function used for the fitting, i.e. the mean measure:
f(C,D)＝g(C)/d2(C,D)
wherein g (C) represents the arc length of curve C, d2(C, D) is the one-way modified Hausdorff distance from C to D:
wherein C issIs a set of points uniformly sampled from C with very little resolution, | CSAnd | represents the number of points in the point set, and | p-q | represents the modulus of the vector. The optimization problem shown above (i.e., the model fitting problem) is solved using a rhododendron search algorithm. Finally, the 15 candidate curves are compared to obtain an optimal curve
Wherein
The cuckoo search algorithm requires pre-specifying the value range of each variable in the model. Position variable x0、y0And z0The value range of (a) is the bounding box of the data. The arc length variable L ranges from [0,200 ]]. The ranges for the other variables are shown in Table 3. The resolution of the sampling point from the curve is set to 0.3 times the data resolution. In this embodiment, the data resolution is 0.2m, since the original data (i.e. shell points) are first put under the resolution of 0.2mSampling to remove data redundancy. That is, the resolution of sampling points from the curve is set to 6 cm. In summary, the curve reconstruction process leaves only one parameter of convergence tolerance that needs to be fine-tuned in the implementation.
TABLE 3 value ranges of highway curve model variables determined according to highway geometric design criteria
| Variables of | α0 | s0 | rV0 | β | γ | rH |
| Maximum value | 2π | 0.08 | 20000 | 0.05 | 0.0025 | 10000 |
| Minimum value | 0 | -0.08 | 100 | 0 | 0 | 20 |
Fig. 3 shows a reconstructed three-dimensional road curve, and referring to fig. 3, the reconstructed curve in this embodiment is a curve specified by a road geometric design standard, and is suitable for not only digital mapping but also road construction quality evaluation.
The road curve method can automatically and robustly identify the type of the road curve and accurately estimate the parameters of the three-dimensional road curve meeting the geometric design standard of the road under the condition that data contains a large number of outliers, and can be applied to various applications such as digital mapping, automatic driving, geometric design of the road, traffic accident analysis, road construction quality evaluation and the like.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. A three-dimensional road curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud is characterized by comprising the following steps: the method comprises the following steps:
s1, dividing the vehicle-mounted mobile laser scanning original point cloud into a plurality of sub-point clouds by using the driving track information of the laser scanning vehicle;
s2, carrying out road surface extraction on each section of sub-point cloud obtained in the step S1 to obtain a road surface point set;
s3, extracting road signs from the road surface point set obtained in the step S2 to obtain a road sign point set;
s4, clustering the road sign point sets obtained in the step S3 to obtain a plurality of road sign sub-point sets;
s5, performing shell extraction on each road sign sub-point set obtained in the step S4 to obtain a shell point set;
s6, performing model fitting on 15 basic three-dimensional road curve models specified by the road geometric design standard and the shell point set obtained in the step S5 to obtain 15 candidate curves; the specific implementation process of the step S6 is as follows: model fitting is respectively carried out on 15 basic three-dimensional road curve models specified by road geometric design standards and the shell point set obtained in the step S5 to obtain 15 candidate curves, and for each curve model m ∈ {1,2, …,15}, the model fitting method tries to find the following candidate curves:
wherein C ismIs the set of curves defined by the model m, D is the set of data points, f (C, D) is the objective function used for the fitting, i.e. the average measure:
f(C，D)＝g(C)/d2(C，D)
wherein g (C) represents the arc length of curve C, d2(C, D) is the one-way modified Hausdorff distance from C to D:
wherein C issIs a set of points uniformly sampled from C with very little resolution, | CSThe model fitting problem shown above is solved by using a rhododendron search algorithm;
s7, comparing the candidate curves obtained in the step S6 to obtain an optimal curve, wherein the optimal curve is
Wherein
2. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 1, wherein the method comprises the following steps: the specific process of step S1 is as follows: inputting a vehicle-mounted mobile laser scanning original point cloud G and a track point cloud T corresponding to the original point cloud G, firstly dividing the T into J subsets according to the storage sequence of points: t is1，T2，…，TJAfter segmentation, except for the last subset TJEach of the remaining subsets has E points, and then G is divided into J segments G as follows1，G2，…，GJWherein
Where d (u, v) represents the Euclidean distance between two three-dimensional points, R ∈ R is the distance threshold used to determine the neighborhood of points, G1And T1Corresponding 1 st set of points, G, divided from GjAnd TjCorresponding j-th set of points, T, divided from GjIs the jth subset divided from Tj-1Is the j-1 th subset divided from T.
3. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 2, wherein the method comprises the following steps: the distance threshold r for determining the neighborhood of points is 10 meters.
4. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 1, wherein the method comprises the following steps: the specific process of step S2 is: setting a smoothness threshold, extracting the road surface by using a smoothness-based region growing algorithm, obtaining a plurality of sub-point sets with different sizes after region growing, and considering the sub-point set with the largest number of points as the extracted road surface point set.
5. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 4, wherein the method comprises the following steps: the smoothness threshold is set to 2 degrees.
6. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 1, wherein the method comprises the following steps: the implementation process of step S3 includes: an intensity variance threshold is set and a 0tsu threshold algorithm based on intensity variance is used to extract road signs.
7. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 1, wherein the method comprises the following steps: in step S4, the road sign point sets extracted in step S3 are clustered by using euclidean clustering to obtain a plurality of road sign sub-point sets.
8. The method for reconstructing the three-dimensional road curve based on the vehicle-mounted mobile laser scanning point cloud as claimed in claim 1, wherein the method comprises the following steps: in step S5, the shell point set is extracted from the road sign subset obtained in step S4 by the alpha shape method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910674347.7A CN110379007B (en) | 2019-07-25 | 2019-07-25 | Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910674347.7A CN110379007B (en) | 2019-07-25 | 2019-07-25 | Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110379007A CN110379007A (en) | 2019-10-25 |
| CN110379007B true CN110379007B (en) | 2022-01-18 |
Family
ID=68255750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910674347.7A Active CN110379007B (en) | 2019-07-25 | 2019-07-25 | Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110379007B (en) |
Families Citing this family (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112880693B (en) * | 2019-11-29 | 2024-07-19 | 北京市商汤科技开发有限公司 | Map generation method, positioning method, device, equipment and storage medium |
| CN111710039B (en) * | 2020-06-03 | 2024-06-14 | 纵目科技（上海）股份有限公司 | High-precision map construction method, system, terminal and storage medium |
| CN112733696B (en) * | 2021-01-04 | 2023-08-15 | 长安大学 | Vehicle-mounted LIDAR road edge extraction method based on multi-model fitting |
| CN112767429B (en) * | 2021-01-18 | 2022-11-01 | 南京理工大学 | Ground-snow surface point cloud rapid segmentation method |
| CN112989946B (en) * | 2021-02-10 | 2024-03-19 | 福瑞泰克智能系统有限公司 | Lane line determining method, device, equipment and vehicle |
| TWI767601B (en) * | 2021-03-10 | 2022-06-11 | 廣達電腦股份有限公司 | Electronic device and method for indoor positioning, imaging, detecting, posture discrimination, and shape discrimination |
| CN116153057A (en) * | 2022-09-12 | 2023-05-23 | 东北林业大学 | Method for estimating lane width based on laser radar point cloud |
| CN118135123B (en) * | 2024-05-10 | 2024-07-12 | 兰州交通大学 | Reverse modeling method, device, equipment, medium and product for expressway |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104766058A (en) * | 2015-03-31 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining lane line |
| CN105404844A (en) * | 2014-09-12 | 2016-03-16 | 广州汽车集团股份有限公司 | Road boundary detection method based on multi-line laser radar |
| CN104376595B (en) * | 2014-11-28 | 2017-03-29 | 史文中 | A kind of three-dimensional road generation method cooperateed with based on airborne LiDAR and GIS |
| CN108845569A (en) * | 2018-04-27 | 2018-11-20 | 厦门维斯云景信息科技有限公司 | Generate semi-automatic cloud method of the horizontal bend lane of three-dimensional high-definition mileage chart |
| CN109858460A (en) * | 2019-02-20 | 2019-06-07 | 重庆邮电大学 | A kind of method for detecting lane lines based on three-dimensional laser radar |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108345822B (en) * | 2017-01-22 | 2022-02-01 | 腾讯科技（深圳）有限公司 | Point cloud data processing method and device |
| WO2018205119A1 (en) * | 2017-05-09 | 2018-11-15 | 深圳市速腾聚创科技有限公司 | Roadside detection method and system based on laser radar scanning |
-
2019
- 2019-07-25 CN CN201910674347.7A patent/CN110379007B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105404844A (en) * | 2014-09-12 | 2016-03-16 | 广州汽车集团股份有限公司 | Road boundary detection method based on multi-line laser radar |
| CN104376595B (en) * | 2014-11-28 | 2017-03-29 | 史文中 | A kind of three-dimensional road generation method cooperateed with based on airborne LiDAR and GIS |
| CN104766058A (en) * | 2015-03-31 | 2015-07-08 | 百度在线网络技术（北京）有限公司 | Method and device for obtaining lane line |
| CN108845569A (en) * | 2018-04-27 | 2018-11-20 | 厦门维斯云景信息科技有限公司 | Generate semi-automatic cloud method of the horizontal bend lane of three-dimensional high-definition mileage chart |
| CN109858460A (en) * | 2019-02-20 | 2019-06-07 | 重庆邮电大学 | A kind of method for detecting lane lines based on three-dimensional laser radar |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110379007A (en) | 2019-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110379007B (en) | Three-dimensional highway curve reconstruction method based on vehicle-mounted mobile laser scanning point cloud | |
| Zai et al. | 3-D road boundary extraction from mobile laser scanning data via supervoxels and graph cuts | |
| CN111179152B (en) | Road identification recognition method and device, medium and terminal | |
| Holgado‐Barco et al. | Automatic inventory of road cross‐sections from mobile laser scanning system | |
| Fathi et al. | Detecting road intersections from GPS traces | |
| Yao et al. | Extraction and motion estimation of vehicles in single-pass airborne LiDAR data towards urban traffic analysis | |
| Ma et al. | Generation of horizontally curved driving lines in HD maps using mobile laser scanning point clouds | |
| CN110197173B (en) | Road edge detection method based on binocular vision | |
| CN112070756B (en) | Three-dimensional road surface disease measuring method based on unmanned aerial vehicle oblique photography | |
| US11941878B2 (en) | Automated computer system and method of road network extraction from remote sensing images using vehicle motion detection to seed spectral classification | |
| CN106096497B (en) | A kind of house vectorization method for polynary remotely-sensed data | |
| Zhang et al. | 3D highway curve reconstruction from mobile laser scanning point clouds | |
| CN111914817B (en) | Road identification method and system based on seed points | |
| Mhangara et al. | Road extraction using object oriented classification | |
| Liu et al. | Image-translation-based road marking extraction from mobile laser point clouds | |
| CN107463944B (en) | A kind of road information extracting method using multidate High Resolution SAR Images | |
| Chen et al. | Reconstruction of building models with curvilinear boundaries from laser scanner and aerial imagery | |
| Omidalizarandi et al. | Segmentation and classification of point clouds from dense aerial image matching | |
| Xu et al. | A method of 3d building boundary extraction from airborne lidar points cloud | |
| CN106709473B (en) | Voxel-based airborne LIDAR road extraction method | |
| TW202228091A (en) | Method, data processing apparatus and computer program product for generating map data | |
| Ma et al. | Automated extraction of driving lines from mobile laser scanning point clouds | |
| Guan | Automated extraction of road information from mobile laser scanning data | |
| CN113807272A (en) | Method for rapidly extracting disaster situations of disaster damaged roads on remote sensing platform based on vector guidance | |
| Zhang | Photogrammetric point clouds: quality assessment, filtering, and change detection |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240205 Address after: Room A802, No. 1819, Luling Road, Siming District, Xiamen, Fujian 361008 Patentee after: XIAMEN KINGTOP INFORMATION TECHNOLOGY Co.,Ltd. Country or region after: China Address before: Siming District of Xiamen city in Fujian Province, 361000 South Siming Road No. 422 Patentee before: XIAMEN University Country or region before: China |
|
| TR01 | Transfer of patent right |