CN112762946B - Road segmentation processing method and system - Google Patents
Road segmentation processing method and system Download PDFInfo
- Publication number
- CN112762946B CN112762946B CN202011504231.8A CN202011504231A CN112762946B CN 112762946 B CN112762946 B CN 112762946B CN 202011504231 A CN202011504231 A CN 202011504231A CN 112762946 B CN112762946 B CN 112762946B
- Authority
- CN
- China
- Prior art keywords
- azimuth
- sequence
- road
- jump
- subsequence
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C21/00—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
- G01C21/26—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network
- G01C21/28—Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00 specially adapted for navigation in a road network with correlation of data from several navigational instruments
- G01C21/30—Map- or contour-matching
- G01C21/32—Structuring or formatting of map data
Landscapes
- Engineering & Computer Science (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method and a system for processing road segments, which are used for extracting a point row for representing a road center line and carrying out linear interpolation, and the road center line is segmented into a plurality of road segments by utilizing the point row after the linear interpolation; sequentially calculating azimuth jump variables of each road segment relative to the previous road segment, and constructing an azimuth jump quantity sequence; dividing the azimuth jump quantity sequence into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon, replacing azimuth jump variables in the subsequences by using a subsequence jump variable average value, and establishing an approximate azimuth jump quantity sequence; according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, the subsequences are collected and divided, and then the road segments are segmented and divided; and fitting the road segment set corresponding to each sub-sequence into a straight line segment or a curve segment according to the azimuth jump variable mean value of each sub-sequence. The scheme can effectively reduce cost and improve the digitizing efficiency.
Description
Technical Field
The invention relates to the technical field of high-precision map making, in particular to a method and a system for processing road segments in a navigation map or high-precision map making process.
Background
In the field of navigation maps and high-precision maps, the segmentation processing of roads or lines, such as laser point clouds or vehicle running tracks obtained by mapping vehicles and performing field collection, is often encountered, and in the case of internal processing, the storage of data is inconvenient for an excessively long road, and the segmentation processing is a necessary process when curve fitting is performed on the road or the line, curvature is calculated, and the like. The current common practice is: the method has the advantages that the cost is high and the efficiency is low by manually calibrating the segmented positions during the process of collecting the control points.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a method and a system for road segmentation processing, which are used for carrying out segmentation processing on a road or a vehicle line by combining the space geometric characteristics of the road and the vehicle line. Because the segmentation position does not need to be manually calibrated, the cost can be effectively reduced, and the digitizing efficiency can be improved.
The technical scheme for solving the technical problems is as follows:
in one aspect, the present invention provides a method for road segment processing, including the steps of:
extracting a point row for representing the road center line, performing linear interpolation, and dividing the road center line into a plurality of road segments by using the point row after the linear interpolation;
sequentially calculating azimuth jump variables of each road segment relative to the previous road segment, and constructing an azimuth jump quantity sequence, wherein the azimuth is a deflection angle of the road segment relative to the reference direction;
dividing the azimuth jump quantity sequence into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon, replacing azimuth jump variables in the subsequences by using a subsequence jump variable average value, and establishing an approximate azimuth jump quantity sequence;
according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, the subsequences are collected and divided, and then the road segments are segmented and divided;
and fitting the road segment set corresponding to each sub-sequence into a straight line segment or a curve segment according to the azimuth jump variable mean value of each sub-sequence.
Further, the extracting the point sequence for representing the center line of the road and performing linear interpolation includes:
and setting a fixed interval M, and inserting new points into the original road center line point row according to the interval M, so as to form a new road center line point row.
Further, dividing the azimuth hopping amount sequence into a plurality of sub-sequences according to a preset hopping amount approximate threshold epsilon, including:
defining a dividing condition: the serial numbers of each azimuth jump variable in the divided subsequence in the original azimuth jump quantity sequence are continuous, and the difference value between the maximum value and the minimum value of the azimuth jump variable in the subsequence is not more than epsilon;
the azimuth jump variable satisfying the division condition is divided into one sub-sequence.
Further, the preset segmentation minimum point threshold delta is obtained according to the following method:
and determining the shortest distance S of the segments according to the condition of the actual road, and after linear interpolation, setting the center point spacing of the center line point row of the road to be M, and then setting the threshold value delta=S/M of the minimum points of the segments.
Further, the classifying and dividing the subsequences according to the preset segmentation minimum point threshold delta, the length of each subsequence and the front and rear subsequences, and further performing segmentation division on the road segments, including:
traversing the approximate azimuth hopping sequence;
if the length of the sub-sequence is not less than the threshold delta, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the lengths of the continuous subsequences are smaller than the threshold delta, the continuous subsequences are gathered, and whether the road segments corresponding to the combined subsequences are divided into independent road segments is judged according to the length of the combined subsequences and the position of the combined subsequences in the road center line point column.
Further, the determining whether the road segment corresponding to the merged sub-sequence is divided into independent road segments according to the length of the merged sub-sequence and the position of the merged sub-sequence in the road center line point column includes:
if the length of the sub-sequence after the gathering is not less than the threshold delta, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the length of the sub-sequence after the gathering is smaller than the threshold delta and the sub-sequence is positioned at the beginning or the end of the road center line point column, judging whether the road segment corresponding to the sub-sequence is used as an independent road segment to be divided according to the azimuth jump variable average value of the sub-sequence and the adjacent sub-sequence;
if the length of the sub-sequence after the collection is smaller than the threshold delta and the sub-sequence is positioned in the middle of the road center line point column, judging whether the road segment corresponding to the sub-sequence is used as an independent road segment to be divided according to the azimuth jump quantity average value of the front sub-sequence and the rear sub-sequence.
Further, the determining whether to divide the road segment corresponding to the subsequence as an independent road segment according to the average value of the azimuth jump variables of the subsequence and the adjacent subsequence includes:
if the azimuth jump variable average value of the subsequence is 0 or the azimuth jump variable average value of the subsequence is not 0 and the azimuth jump variable average value of the adjacent subsequence is 0, dividing the road segment corresponding to the subsequence as an independent road segment;
if the average value of azimuth jump variables of the subsequence and the adjacent subsequence is not 0, merging the subsequence into the adjacent subsequence, and dividing the road segment corresponding to the merged subsequence as an independent road segment;
further, the determining whether to divide the road segment corresponding to the subsequence as an independent road segment according to the average value of the azimuth jump amounts of the front and rear subsequences includes:
if the azimuth jump quantity mean values of the front sub-sequence and the rear sub-sequence are the same, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the azimuth jump variable mean values of the front subsequence and the rear subsequence are different, merging the subsequences into subsequences with the azimuth jump variable mean values being closer, and dividing road segments corresponding to the merged subsequences into independent road segments.
Further, the fitting the road segment corresponding to each segment sequence into a straight line segment or a curve segment according to the segmentation result and the azimuth jump variable mean value of each segment sequence includes:
the azimuth jump variable of each point in the sequence corresponding to the road segment is 0, and the road segment is fitted into a straight line segment;
the azimuth jump quantity of each point of the sequence corresponding to the road segment is constant and is not 0, and the road segment is fitted into a circular curve segment;
and if the azimuth jump variable of each point in the sequence corresponding to the road segment is not 0 and is not a constant value, fitting the road segment into a moderation curve segment.
On the other hand, the invention also provides a road segmentation processing system, which comprises:
the preprocessing module is used for extracting a point row for representing the central line of the road and performing linear interpolation; calculating an angle of the road segment separated by each point in the point row with respect to the reference direction, defined as an azimuth;
the scatter diagram construction module is used for calculating azimuth angle jump variables of each point in the point column after linear interpolation relative to the previous point, and constructing a scatter diagram, wherein the abscissa of the scatter diagram is the point number in the point column after linear difference, and the ordinate is the jump variable of each point azimuth angle;
the scatter diagram updating module is used for dividing points in the scatter diagram into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon and a numbering sequence, replacing jump variables of each point in the subsequences by jump quantity average values of all points in the subsequences, establishing an approximate azimuth jump quantity sequence and updating the scatter diagram;
the road segment dividing module is used for carrying out aggregation and division on the subsequences according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, and carrying out segmentation division on the road segments;
and the fitting module is used for fitting the road segments corresponding to the subsequences into straight line segments or curve segments according to the segmentation result and the azimuth jump variable average value of the subsequences.
The beneficial effects of the invention are as follows: (1) the general idea of road engineering design is referred, the segmentation processing is closer to reality, and the subsequent road digital processing is facilitated. Such as: smoothing processing, calculating the curvature of a road or a vehicle line, and the like are performed using a curve fitting method.
(2) The method only depends on the space geometric characteristic data of the processing object (road or vehicle line), and the segmentation method has strong universality and is convenient for automatic processing.
Drawings
FIG. 1 is a flowchart of a road segmentation processing method according to an embodiment of the present invention;
FIG. 2 is a schematic view of a point array for characterizing a centerline of a roadway to be segmented according to an embodiment of the present invention;
FIG. 3 is a scatter diagram of azimuth jump variables for each point of a road to be segmented according to an embodiment of the present invention;
FIG. 4 is a scatter diagram after the azimuth jump variable approximation process provided by the embodiment of the invention, and (5) rectangular frames are sequences to be combined;
fig. 5 is a schematic view of a segmented road according to an embodiment of the present invention.
Detailed Description
The principles and features of the present invention are described below with reference to the drawings, the examples are illustrated for the purpose of illustrating the invention and are not to be construed as limiting the scope of the invention.
Because road construction meets certain engineering technical standards, the digital road construction meets the standards. When the road is constructed, the road is generally composed of three forms or the combination of the three forms, which are respectively: (1) straight line, (2) circular curve, (3) relaxation curve. These three curves have unique geometric features. The invention combines the space geometric characteristics of the road and the vehicle line according to the characteristic to obtain a simple and feasible segmentation method. Because the segmentation position does not need to be manually calibrated, the cost can be effectively reduced, and the digitizing efficiency can be improved.
Specifically, an embodiment of the present invention provides a road segment processing system, including:
the preprocessing module is used for extracting a point row for representing the central line of the road and carrying out linear interpolation, and dividing the central line of the road into a plurality of road segments by utilizing the point row after the linear interpolation;
the azimuth jump quantity sequence construction module is used for sequentially calculating azimuth jump variables of each road segment relative to the previous road segment and constructing an azimuth jump quantity sequence, wherein the azimuth is a deflection angle of the road segment relative to the reference direction;
the approximate azimuth jump quantity sequence construction module is used for dividing the azimuth jump quantity sequence into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon, and replacing each azimuth jump variable in the subsequences by using a subsequence jump variable average value to establish an approximate azimuth jump quantity sequence;
the road segment repartition module is used for collecting and dividing the subsequences according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, and further carrying out segmentation division on the road segments;
and the fitting module is used for fitting the road segment set corresponding to each subsequence into a straight line segment or a curve segment according to the azimuth jump variable mean value of each subsequence.
The invention refers to the general idea of road engineering design, and the segmentation processing is closer to reality, thereby being more beneficial to the subsequent road digital processing. Such as: smoothing processing, calculating the curvature of a road or a vehicle line, and the like are performed using a curve fitting method.
The scheme only depends on the space geometric characteristic data of the processing object (road or vehicle line), and the segmentation method is strong in universality and convenient for automatic processing.
With the above-mentioned road segment processing system, the embodiments of the present invention include the following aspects when performing road segment processing:
1. road center line acquisition and pretreatment
The process is to first obtain the road data to be segmented and then linearly interpolate the obtained road data. The method specifically comprises the following two aspects:
1. acquisition of road center line
The road data, i.e., the point row P of the road center line, forms a specific geometric shape on a two-dimensional plane.
2. Pretreatment of
Considering that the obtained point columns of the road center line may have uneven distribution, which is not beneficial to subsequent processing, a fixed interval M is set, and the point columns of the road center line are readjusted according to the fixed interval by a linear interpolation method.
(1) Distance M between set points
The setting of the pitch M of the dots can be referred to the expressway construction standard, the value of which must not be higher than the minimum road length required for every 1 ° of deflection of the expressway (excluding the ramp or the expressway connection).
For roads with speed limit below 60km/s, the road conditions are generally not ideal, and the distance M between points may be suitably small in consideration of the accuracy of calculation, for example: m=5 meters.
For roads with speed limits of 60km/s and above, the distance M between points may be suitably larger in view of reducing the calculation amount, for example, because the roads are relatively straight: m=10 meters.
(2) Linear interpolation
At a fixed pitch M, points are inserted on the original sequence P to form a new sequence P', as shown in fig. 2. Denoted as P' = [ P ] 1 ，p 2 ，p 3 ……p m ]M is the number of midpoints in the dot column after linear interpolation.
2. Road segmentation processing
The process performs segmentation processing on the road. The method mainly comprises the following steps:
1. azimuthal angle calculation
Calculating the center line segment P of each road according to the P' sequence i p i+1 Angle a relative to the north direction (i.e., the Y-axis positive direction) i (1.ltoreq.i.ltoreq.m-1), defined as azimuth angle.
2. Constructing a hopping sequence of azimuth angles to form a scatter diagram
Sequentially calculating the jump quantity da of azimuth angles of two adjacent road segments i ＝a i+1 -a i I.e. the azimuthal jump amount, the azimuthal jump amount sequence is recorded as da= [ dA ] 1 ,da 2 ,da 3 ,...,da k ](k＝m-2)。
Taking dA sequence number i as a value on an X axis, and taking the jump quantity dA of azimuth angles of each point as a value i For values on the Y-axis, a scatter plot of the azimuth jump variable sequence is created. For example, as shown in fig. 3.
3. Azimuthal jump quantity approximation processing
(1) Setting a jump variable approximation processing threshold epsilon
The position of the road centerline point may deviate when the road data is collected. To eliminate the deviation, it is conceivable to set a jump variable approximation processing threshold epsilon.
For roads with speed limit below 60km/s, the azimuth jump variable approximation threshold ε may be set to a small value (e.g., 2, i.e., the azimuth float value is + -1 °) because the point spacing M is small.
For roads with speed limits of 60km/s and above, the azimuth jump variable approximation threshold ε may be set to a large value (e.g., 4, i.e., the azimuth float value is + -2 °) because the point spacing M is relatively large.
(2) Jump quantity approximation processing
From the hop amount sequence dA obtained in procedure 2, starting from the 1 st hop amount
a) Continuously taking the value da in the hop sequence i Forming a subsequence.
Marked as dA j ＝[da j1 ，da j2 ，da j3 ……da jo ](o≤m-2)
This subsequence satisfies the following two conditions.
(1) The sub-sequence satisfies that the difference between the maximum value of the transition amount and the minimum value of the transition amount is not greater than a threshold epsilon. Namely: max (dA) j )-min(dA j )≤ε；
(2) The length is longest.
b) All hop magnitudes for this sub-sequence are replaced by the average value for the sub-sequence.
Namely: da A ji ＝average(dA j )(1≤i≤o)
c) And c) repeating the operations a) and b) in sequence until all the sequences are processed. A scatter plot of the approximate azimuth hop variable sequence is created as shown in fig. 4.
4. Azimuthal hopping sequence segmentation processing
And analyzing the form of each subsequence according to the minimum point number threshold value of the segmentation, and carrying out segmentation judgment on the azimuth angle jump variable sequence.
The specific process is as follows:
(1) Determining a segment minimum points threshold delta
And determining the shortest distance S of the segments according to the condition of the actual road, and knowing the distance M of points, thereby obtaining the threshold delta=S/M of the minimum number of the segments.
The segmentation processing is used for distinguishing road forms, such as a linear type or a curve type, and different settings are considered according to the speed limit condition of the road in order to balance the relation between the segmentation accuracy and the calculated amount. For example:
for roads with speed limit below 60km/S, in case of non-ideal road conditions, S may be set to a distance (denoted as A, for example: 30 meters) that ensures that the vehicle needs to travel at the intersection for normal turns. S=a=30 meters.
For roads with speed limit of 60km/S and above, S can be set as a multiple of A in order to reduce the calculation amount, for example: s= 3*A =3×30=90 meters.
Segment minimum point threshold δ=s/M. For example:
roads with speed limit below 60 km/s: δ=30/5=6
Road with speed limit of 60km/s and above, delta=90/10=9
(2) Sub-sequence aggregation and partitioning
And carrying out aggregation and division on the subsequences according to a certain rule and carrying out segmentation judgment according to the lengths of each subsequence and the front and rear subsequences.
The specific process is as follows:
(1) traversing approximate azimuth hopping quantum sequence dA j 。
(2) If the subsequence dA j The length of the sub-sequence is not less than the threshold delta, and the road segment corresponding to the sub-sequence is used as independent road segment division.
(3) If a plurality of subsequences dA occur consecutively j Fragments of length less than threshold delta are combined and a plurality of consecutive subsequences are considered as a new subsequence.
(4) If the length of the subsequence after the collection processing in the step (3) is not less than the threshold delta, the road segment corresponding to the subsequence is used as an independent road segment division.
(5) If the length of the sub-sequence formed after the (3) aggregation process is less than the threshold delta, and the road segment corresponding to the sub-sequence is located at the beginning or end of the segmented road,
a) If the average value of the azimuth jump variables of the subsequence is 0, the road segment corresponding to the subsequence is divided into independent road segments.
b) If the azimuth jump variable mean value of the subsequence is not 0, and the adjacent subsequence is also not 0, merging and collecting the subsequence and the adjacent subsequence, and then dividing the corresponding road segment as an independent road segment. If the average value of the azimuth jump variables of the adjacent sequences is 0, the road segments corresponding to the subsequences are used as independent road segment partitions.
(6) If the length of the sub-sequence formed after the collection processing in the step (3) is smaller than the threshold delta, and the road segment corresponding to the sub-sequence is positioned in the middle of the segmented road.
a) If the azimuth jump quantity mean values of the front sub-sequence and the rear sub-sequence are the same, the road segments corresponding to the sub-sequences are divided into independent road segments.
b) If the average value of the azimuth jump amounts of the front subsequence and the rear subsequence is different, the average value of the collected subsequence is compared with the average value of the front subsequence and the rear subsequence respectively, the sequences with the average value being closer to each other are combined, and the road segments corresponding to the combined subsequences are used as new road segment division.
Repeating the steps (1) - (6) until all sub-sequence processing is completed.
Recording the starting and stopping points of the segments divided as independent road segments.
Fragments (1), (2), (3), (4), (5), (6), (7) as indicated in FIG. 4,
wherein the fragment (5) is a fragment after the merging and collecting treatment.
3. Saving segmentation results and post-processing
And D, performing segmentation-based digital processing according to the starting and stopping points of the segments which are obtained in the step two and are divided into independent road segments.
1. Segment result preservation and labeling
(1) The mean value is constant at 0: marked as straight line segment
Because the jump amounts of the azimuth angles are all 0, the road section can be judged to be a straight line section.
For example, the sections (1), (3) and (7) in fig. 5.
(2) The mean value is constant, but not 0: marked as a circular curve segment
Since the jump amount of the azimuth angle is not 0 and is a constant value, the road section can be judged to be a circular curve section.
For example, the sections (2), (4) and (6) in fig. 5.
(3) The mean value is not constant: marked as a gentle curve segment
Since the transition amount of the azimuth angle is not 0 and is not a constant value, there is a case where the transition amount of the azimuth angle gradually increases or gradually decreases, and it can be determined that the road section is a gentle curve section.
For example, as shown in section (5) of fig. 5.
2. Post-treatment
(1) Smoothing process
a) The portion marked as a straight line segment:
since this portion has typical linear characteristics, the segment can be smoothed directly using a linear fitting method.
b) The portion marked as a circular curve segment:
since this portion has the characteristic of a typical circle, the segment can be smoothed directly using a circular curve fitting method.
c) The portion labeled as a mild curve segment:
since this portion is characteristic of a typical relaxation curve, the segment can be smoothed using a fitting method of a parabola, a clothoid curve, a polynomial, or the like.
(2) Curvature attribute calculation
a) The portion marked as a straight line segment: curvature is set to 0
b) The portion marked as a circular curve segment: the curvature is set to a fixed value (1/R)
c) The portion labeled as a mild curve segment: and calculating a curvature value according to a standard curvature formula according to the fitted curve equation, and setting.
The foregoing description of the preferred embodiments of the invention is not intended to limit the invention to the precise form disclosed, and any such modifications, equivalents, and alternatives falling within the spirit and scope of the invention are intended to be included within the scope of the invention.
Claims (9)
1. A method of road segmentation processing, comprising the steps of:
extracting a point row for representing the road center line, performing linear interpolation, and dividing the road center line into a plurality of road segments by using the point row after the linear interpolation;
sequentially calculating azimuth jump variables of each road segment relative to the previous road segment, and constructing an azimuth jump quantity sequence, wherein the azimuth is a deflection angle of the road segment relative to the reference direction;
dividing the azimuth jump quantity sequence into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon, replacing azimuth jump variables in the subsequences by using a subsequence jump variable average value, and establishing an approximate azimuth jump quantity sequence;
according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, the subsequences are collected and divided, and then the road segments are segmented and divided;
according to the azimuth jump variable mean value of each sub-sequence, fitting the road segment set corresponding to each sub-sequence into a straight line segment or a curve segment;
dividing the azimuth hopping amount sequence into a plurality of subsequences according to a preset hopping amount approximate threshold epsilon, wherein the method comprises the following steps:
defining a dividing condition: the serial numbers of each azimuth jump variable in the divided subsequence in the original azimuth jump quantity sequence are continuous, and the difference value between the maximum value and the minimum value of the azimuth jump variable in the subsequence is not more than epsilon;
the azimuth jump variable satisfying the division condition is divided into one sub-sequence.
2. The method of claim 1, wherein extracting and linearly interpolating the array of points characterizing the centerline of the roadway comprises:
and setting a fixed interval M, and inserting new points into the original road center line point row according to the interval M, so as to form a new road center line point row.
3. The method of claim 1, wherein the predetermined segmentation minimum point threshold δ is obtained by:
and determining the shortest distance S of the segments according to the condition of the actual road, and after linear interpolation, setting the center point spacing of the center line point row of the road to be M, and then setting the threshold value delta=S/M of the minimum points of the segments.
4. A method according to claim 1 or 3, wherein the classifying and dividing the subsequences according to the preset segmentation minimum point threshold δ, the length of each subsequence and the preceding and following subsequences, and further performing segmentation division on the road segments, includes:
traversing the approximate azimuth hopping sequence;
if the length of the sub-sequence is not less than the threshold delta, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the lengths of the continuous subsequences are smaller than the threshold delta, the continuous subsequences are gathered, and whether the road segments corresponding to the combined subsequences are divided into independent road segments is judged according to the length of the combined subsequences and the position of the combined subsequences in the road center line point column.
5. The method of claim 4, wherein the determining whether the road segment corresponding to the merged sub-sequence is divided into independent road segments according to the length of the merged sub-sequence and the position of the merged sub-sequence in the road centerline point row comprises:
if the length of the sub-sequence after the gathering is not less than the threshold delta, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the length of the sub-sequence after the gathering is smaller than the threshold delta and the sub-sequence is positioned at the beginning or the end of the road center line point column, judging whether the road segment corresponding to the sub-sequence is used as an independent road segment to be divided according to the azimuth jump variable average value of the sub-sequence and the adjacent sub-sequence;
if the length of the sub-sequence after the collection is smaller than the threshold delta and the sub-sequence is positioned in the middle of the road center line point column, judging whether the road segment corresponding to the sub-sequence is used as an independent road segment to be divided according to the azimuth jump quantity average value of the front sub-sequence and the rear sub-sequence.
6. The method of claim 5, wherein the determining whether to divide the road segment corresponding to the sub-sequence as an independent road segment according to the average value of the azimuth jump variables of the sub-sequence and the neighboring sub-sequence comprises:
if the azimuth jump variable average value of the subsequence is 0 or the azimuth jump variable average value of the subsequence is not 0 and the azimuth jump variable average value of the adjacent subsequence is 0, dividing the road segment corresponding to the subsequence as an independent road segment;
if the average value of the azimuth jump variables of the subsequence and the adjacent subsequence is not 0, merging the subsequence into the adjacent subsequence, and dividing the road segment corresponding to the merged subsequence as an independent road segment.
7. The method of claim 5, wherein the determining whether to divide the road segment corresponding to the sub-sequence as an independent road segment according to the average value of the azimuth jump amounts of the front and rear sub-sequences comprises:
if the azimuth jump quantity mean values of the front sub-sequence and the rear sub-sequence are the same, dividing the road segment corresponding to the sub-sequence as an independent road segment;
if the azimuth jump variable mean values of the front subsequence and the rear subsequence are different, merging the subsequences into subsequences with the azimuth jump variable mean values being closer, and dividing road segments corresponding to the merged subsequences into independent road segments.
8. The method of claim 1, wherein fitting the road segment corresponding to each segment sequence to a straight line segment or a curved line segment based on the segmentation result and the azimuth jump variable mean value of each segment sequence comprises:
the azimuth jump variable of each point in the sequence corresponding to the road segment is 0, and the road segment is fitted into a straight line segment;
the azimuth jump quantity of each point of the sequence corresponding to the road segment is constant and is not 0, and the road segment is fitted into a circular curve segment;
and if the azimuth jump variable of each point in the sequence corresponding to the road segment is not 0 and is not a constant value, fitting the road segment into a moderation curve segment.
9. A road segment processing system, comprising:
the preprocessing module is used for extracting a point row for representing the central line of the road and carrying out linear interpolation, and dividing the central line of the road into a plurality of road segments by utilizing the point row after the linear interpolation;
the azimuth jump quantity sequence construction module is used for sequentially calculating azimuth jump variables of each road segment relative to the previous road segment and constructing an azimuth jump quantity sequence, wherein the azimuth is a deflection angle of the road segment relative to the reference direction;
the approximate azimuth jump quantity sequence construction module is used for dividing the azimuth jump quantity sequence into a plurality of subsequences according to a preset jump quantity approximate threshold epsilon, and replacing each azimuth jump variable in the subsequences by using a subsequence jump variable average value to establish an approximate azimuth jump quantity sequence;
the road segment repartition module is used for collecting and dividing the subsequences according to a preset segmentation minimum point threshold delta, the length of each subsequence and the length of the front subsequence and the length of the rear subsequence, and further carrying out segmentation division on the road segments;
the fitting module is used for fitting the road segment set corresponding to each subsequence into a straight line segment or a curve segment according to the azimuth jump variable mean value of each subsequence;
dividing the azimuth hopping amount sequence into a plurality of subsequences according to a preset hopping amount approximate threshold epsilon, wherein the method comprises the following steps:
defining a dividing condition: the serial numbers of each azimuth jump variable in the divided subsequence in the original azimuth jump quantity sequence are continuous, and the difference value between the maximum value and the minimum value of the azimuth jump variable in the subsequence is not more than epsilon;
the azimuth jump variable satisfying the division condition is divided into one sub-sequence.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011504231.8A CN112762946B (en) | 2020-12-17 | 2020-12-17 | Road segmentation processing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011504231.8A CN112762946B (en) | 2020-12-17 | 2020-12-17 | Road segmentation processing method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112762946A CN112762946A (en) | 2021-05-07 |
| CN112762946B true CN112762946B (en) | 2023-06-27 |
Family
ID=75694958
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011504231.8A Active CN112762946B (en) | 2020-12-17 | 2020-12-17 | Road segmentation processing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112762946B (en) |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1632925A1 (en) * | 2003-06-11 | 2006-03-08 | Matsushita Electric Industrial Co., Ltd. | Digital map position information compressing method and device |
| US20090144031A1 (en) * | 2007-12-04 | 2009-06-04 | Tele Atlas North America, Inc. | System for applying clothoid curve values to roadways in a geographic data information system |
| JP4479817B2 (en) * | 2008-03-28 | 2010-06-09 | アイシン・エィ・ダブリュ株式会社 | Road shape estimation device, road shape estimation method and program |
| EP2105708B1 (en) * | 2008-03-28 | 2013-12-04 | Aisin AW Co., Ltd. | Road shape estimating device, road shape estimating method and program |
| CN106595656B (en) * | 2016-12-29 | 2019-09-10 | 辽宁工程技术大学 | A kind of identification of highway alignment feature and calculation method of parameters based on vehicle-mounted POS |
| CN108286979B (en) * | 2017-01-09 | 2021-07-20 | 北京四维图新科技股份有限公司 | Method and device for acquiring high-precision navigation path data, navigation system and driving system |
| CN108280840B (en) * | 2018-01-11 | 2021-09-03 | 武汉理工大学 | Road real-time segmentation method based on three-dimensional laser radar |
| CN109033176B (en) * | 2018-06-25 | 2024-02-27 | 腾讯科技（北京）有限公司 | Road curvature determination method, device, storage medium and computer equipment |
| CN111198562A (en) * | 2019-12-31 | 2020-05-26 | 武汉中海庭数据技术有限公司 | Preprocessing optimization method for space line characteristics of crowdsourcing fragment map |
-
2020
- 2020-12-17 CN CN202011504231.8A patent/CN112762946B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112762946A (en) | 2021-05-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103778429B (en) | Automatic extraction method for road information in a kind of Vehicle-borne Laser Scanning point cloud | |
| CN110906940B (en) | Lane sideline aggregation method based on track direction | |
| CN110288675B (en) | Lane reference line construction method and device and storage medium | |
| US6560531B1 (en) | Multi-resolution trend metric for shape comparison and applications thereof | |
| CN111796298B (en) | Automatic point cloud point supplementing method for laser LiDAR power line | |
| CN108469263B (en) | Method and system for shape point optimization based on curvature | |
| CN112193239B (en) | Trajectory calculation method, device and storage device in automatic parking control system | |
| CN112762946B (en) | Road segmentation processing method and system | |
| CN111986286A (en) | Radar-based curve drawing method and device, electronic equipment and storage medium | |
| CN117146844A (en) | Method, device, equipment and medium for smoothing and optimizing reference line in path planning | |
| CN111354083A (en) | Progressive building extraction method based on original laser point cloud | |
| CN111504333B (en) | Method and device for adjusting elevation of uplink and downlink roads in three-dimensional map | |
| CN116051841A (en) | Roadside ground object multistage clustering segmentation algorithm based on vehicle-mounted LiDAR point cloud | |
| CN115310522A (en) | Data quality improving method and system for riding track of non-motor vehicle | |
| CN111192216B (en) | Lane line smoothing method and system | |
| CN116433889A (en) | Laser radar target recognition algorithm based on convex plane detection | |
| CN110135219A (en) | Data vacuate method, apparatus, storage equipment, map, control system and vehicle | |
| CN109308468B (en) | Lane line detection method | |
| CN113128307A (en) | Lane line detection processing method, and related device and system | |
| CN114782925B (en) | Highway guardrail vectorization method and device based on vehicle-mounted LIDAR data | |
| CN114114930B (en) | Method, device, equipment and medium for generating local reference path of automobile | |
| CN116360506B (en) | Unmanned aerial vehicle path planning method and system, unmanned aerial vehicle and medium | |
| CN112530183B (en) | City trunk similarity measurement method | |
| CN118036126A (en) | Bisection-based road reconstruction and expansion segmentation method | |
| US20230174099A1 (en) | Method and apparatus for trajectory smoothing in autonomous vehicle control |
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 |