CN116168538A - Planar road vehicle passing space identification method - Google Patents
Planar road vehicle passing space identification method Download PDFInfo
- Publication number
- CN116168538A CN116168538A CN202310154797.XA CN202310154797A CN116168538A CN 116168538 A CN116168538 A CN 116168538A CN 202310154797 A CN202310154797 A CN 202310154797A CN 116168538 A CN116168538 A CN 116168538A
- Authority
- CN
- China
- Prior art keywords
- space
- grid
- road
- passing
- grids
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 230000008859 change Effects 0.000 claims abstract description 9
- 238000004364 calculation method Methods 0.000 claims description 22
- 230000005465 channeling Effects 0.000 claims description 18
- 238000000605 extraction Methods 0.000 claims description 7
- 238000012216 screening Methods 0.000 claims description 7
- 230000001419 dependent effect Effects 0.000 claims description 4
- 238000002372 labelling Methods 0.000 claims description 4
- 230000008569 process Effects 0.000 abstract description 6
- 238000013139 quantization Methods 0.000 abstract 1
- 238000004891 communication Methods 0.000 description 8
- 238000012549 training Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000002955 isolation Methods 0.000 description 4
- 230000008520 organization Effects 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 2
- 238000013480 data collection Methods 0.000 description 2
- 238000003064 k means clustering Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 238000011002 quantification Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 230000002035 prolonged effect Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/29—Geographical information databases
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Remote Sensing (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention provides a plane road vehicle passing space identification method, which is characterized in that the range of a road space to be identified is obtained based on vehicle track data acquired by road front end sensing equipment, and the road space to be identified is divided into a plurality of grids conforming to conditions; then, the road space topological structure, the vehicle passing space boundary, the vehicle passable space and the channelized space are identified by determining the occupied state of the grid, and then the category of the road space is judged according to the road space topological structure, and the area of the vehicle passing space is calculated. The whole identification, classification and quantization process does not need to manually survey and set parameters of the road space in advance, does not need other additional information such as a map and the like, is driven by vehicle track data, ensures objectivity and consistency of the identification result of the vehicle passing space, and can monitor the utilization state and change condition of the vehicle passing space on the plane road in real time.
Description
Technical Field
The invention relates to the technical field of road traffic space recognition, in particular to a plane road vehicle passing space recognition method.
Background
Along with the continuous increase and perfection of the front-end sensing equipment on the road of China, the all-weather traffic state monitoring and traffic data collection of all scenes are basically realized in the urban range. The intelligent and refined management of traffic is realized by utilizing traffic big data enabling front-end sensing equipment, and the intelligent and refined management becomes a new direction of traffic jam relieving technology research in China. In the prior practice, the automatic identification and extraction of intersections in urban road networks are realized based on road traffic space identification technologies, but the granularity of the technologies is relatively large, the judgment of the congestion of the intersections is only stopped on a qualitative macroscopic level, the utilization condition of vehicles in the intersections on the road space cannot be described from the microscopic level, and the further research and judgment of the congestion problem are prevented. The main reason for this problem is that the front-end sensing device cannot obtain and process map space information of the road, and cannot automatically identify the space where vehicles can pass on the road. To solve this problem, researchers try to identify the traffic space of vehicles on the road based on new methods such as machine learning, but although the ideal effect is obtained on the demonstration road, the proposed method or model usually requires manual calibration or assumption of key elements such as shape, spatial range, and discrimination criteria of the road in advance, and requires a background computer to perform calculation training for a long time on a large amount of historical data. The manual data calibration and a large amount of training time can lead to the excessive cost of vehicle passing space identification, and the road vehicle passing space cannot be identified in real time, so that the popularization and popularization of the identification method in practical application are hindered.
Disclosure of Invention
In order to solve the problems of high cost and long time consumption in the prior art for identifying the passing space of the plane road vehicle based on machine learning and other methods, the invention provides the plane road vehicle passing space identification method which has small demand for basic data, does not need manual calibration and machine training, does not depend on a map, has accurate result and high calculation speed, and can monitor the condition of the vehicle using the passing space on the plane road in real time.
The technical scheme of the invention is as follows: a method for identifying a traffic space of a planar road, comprising the steps of:
s1: acquiring vehicle track data to be analyzed according to a preset data acquisition time period and road front end sensing equipment in a specified range;
the method is characterized by further comprising the following steps:
s2: based on the coverage areas of all vehicle track points in the vehicle track data to be analyzed, confirming the space range of the road space to be identified, uniformly dividing the road space to be identified into grids, and marking as: a space grid;
s3: judging the occupation state of each space grid based on the characteristic threshold value for judgment, the coordinate relation of the vehicle track point and the space grid;
The occupancy state of the space grid comprises: occupied and unoccupied;
s4: identifying a road space topological structure in the road space to be identified based on the occupation states of all the space grids in the road space to be identified; the road space topological structure is expressed based on [ road space characteristic values, road space characteristic intervals ];
the method for identifying the road space topological structure comprises the following steps:
a1: the space grids at the outermost side of the road space to be identified form a feature frame, the space grids in the feature frame are traversed clockwise from any space grid in a non-occupied state in the feature frame, the total number of times of occurrence of the space grids in the occupied state is recorded as a feature value of the feature frame, and the feature value is recorded as: a base feature frame feature value;
recording only 1 time when the space grid in the occupied state continuously appears;
if the space grid in the unoccupied state does not exist in the feature frame, marking the feature value of the basic feature frame as 1;
a2: taking the space grids which are directly connected with the space grids forming the feature frame in the road space surrounded by the feature frame, forming a new feature frame, and repeating the step a1 until the space surrounded by the feature frame does not contain any space grids, namely, after all the space grids participate in calculation, obtaining a group of sequences consisting of the feature values of the basic feature frame, and marking the sequences as: a base feature frame feature value sequence;
a3: in the basic feature frame feature value sequence, finding the basic feature frame feature value with highest occurrence frequency is recorded as: calculating the maximum number of continuous occurrence times of the road space characteristic value, and recording the maximum number as: road space feature spacing;
s5: finding all the subsequences of the characteristic values of the basic characteristic frames formed by the characteristic values of the road space in the characteristic value sequence of the basic characteristic frames, wherein the characteristic frame corresponding to the characteristic value of the first basic characteristic frame of the subsequence with the longest length is the traffic space boundary of the vehicle in the road space to be identified;
if the vehicle passing space boundary consists of the space grid at the outermost side of the road space to be identified, selecting the feature frame corresponding to the feature value of the second basic feature frame of the sub-sequence with the longest length as the vehicle passing space boundary;
the space grid in the occupied state in the boundary of the vehicle passing space is marked as: a pass grid; and marking the space grids except the passing grid in the road space to be identified as: a non-passing grid;
s6: screening and re-marking the pass grid and the non-pass grid; according to a connected domain calculation method, the obtained connected domain formed by the passing grids is a vehicle passable space in the road space to be identified;
The occupancy state of the space grid in the vehicle-navigable space is: is occupied;
the occupation state of the space grid which does not belong to the vehicle-passable space in the road space to be identified is as follows: non-occupation;
s7: extracting a channelized grid in the vehicle passing space based on the passing grid and the non-passing grid, wherein a connected domain formed by the channelized grid is marked as: a channeling space;
re-labeling the occupancy state of the space grid within the canalized space as: is occupied;
s8: identifying the road space topology structure in the road space to be identified based on the latest occupied states of all the space grids in the road space to be identified;
s9: the vehicle passing space category is judged according to the road space characteristic value, and the specific method comprises the following steps:
when the road space characteristic value is not more than 2, the vehicle passing space class is a road section;
when the road space characteristic value is equal to 3, the vehicle passing space class is a T-shaped or Y-shaped intersection;
when the road space characteristic value is equal to 4, the vehicle passing space class is a cross or X-shaped road opening;
when the road space characteristic value is greater than 4, the vehicle passing space class is a multi-entrance road junction;
S10: the area of the vehicle passing space is calculated by the following method:
S veh ＝S cell (N veh +N cana )
wherein S is veh For the area of the vehicle passing space S cell For the area of the space grid, N veh For the number, N, of the space grids contained in the vehicle-operable space cana And (3) the number of the space grids contained in the canalized space.
It is further characterized by:
the calculation method of the range of the road space to be identified comprises the following steps:
the left boundary of the range of the road space to be identified is
The right boundary of the range of the road space to be identified is
The upper boundary of the range of the road space to be identified is
The lower boundary of the range of the road space to be identified is
Wherein x is max X is the maximum value of the abscissa in the vehicle position coordinates min Is the minimum value of the abscissa in the vehicle position coordinates, y max Y is the maximum value of the ordinate in the vehicle position coordinates min The minimum value of the ordinate in the vehicle position coordinates is represented by L, and the side length of the space grid is represented by L;
the method for judging the occupancy state of the space grid comprises the following steps:
b1: acquiring space grid coordinates to be judged;
the coordinates of the space grid to be judged are set as follows: left boundary x left The right boundary is x right The upper boundary is y up The lower boundary is y down ；
Initializing a counter for each of the grids to be judged: cg=0;
b2: the vehicle track points in the vehicle track data to be analyzed are acquired one by one and recorded as: vehicle track point to be calculated (x veh ，y veh )；
B3 is implemented on each vehicle track point to be calculated until all the vehicle track points are compared with the space grid to be judged, and step b4 is implemented;
b3: comparing the coordinates of the vehicle track points to be calculated with the coordinates of the space grid to be judged;
if x left ≤x veh <x right And y is left ≤y veh <y right The vehicle track points to be calculated belong to the space grid to be judged, and cg=cg+1;
b4: comparing the obtained Cg with the characteristic threshold for discrimination;
when Cg is greater than or equal to the characteristic threshold for discrimination, the occupancy state of the grid to be discriminated is set as: is occupied;
otherwise, setting to be unoccupied;
in step S6, the specific steps of screening and re-marking the passing grid and the non-passing grid include:
c1: extracting a connected domain formed by the passing grids, and marking the connected domain as: passing through the grid connected domain;
re-marking the passing grid which does not belong to the largest passing grid connected domain as a non-passing grid;
c2: extracting a connected domain consisting of the non-passing grids, and marking the connected domain as: a non-passing grid connected domain;
re-marking the non-passing grid which does not belong to the largest non-passing grid connected domain as a passing grid;
c3: checking the arrangement condition of the passing grids and the non-passing grids row by row and column by column, and re-marking the non-passing grids as passing grids if the two passing grids are separated by one non-passing grid;
c4: the occupancy state of the passing grid is re-marked as: is occupied;
re-marking the occupancy state of the non-passing grid as: non-occupation;
the connected domain extraction method comprises the following steps:
based on a four-communication marking algorithm or an eight-communication marking algorithm, each kind of grids are attributed to one kind of communication domain;
a method of extracting the canalized grid in the vehicle passing space, comprising the steps of:
d1: starting from any one of the grids forming the vehicle passing space boundary, traversing the grids in the vehicle passing space boundary sequentially in the same direction, and recording the number of the non-passing grids when continuously appearing to form: a collection of non-passing grid numbers;
d2: adding elements in the non-passing grid quantity set: 1, a step of;
d3: dividing the number set in the step d2 into two types through a clustering algorithm;
if 1 is singly classified into one type, the canalization grid does not exist, and the algorithm is ended;
otherwise, implementing step d4;
d4: continuously occurring non-passing grids corresponding to the number 1 belonging to the same class are re-labeled: a trenching grid;
d5: extracting a communicating region formed by the non-passing grids in the road space to be identified, and re-marking the non-passing grids which do not belong to the largest non-passing grid communicating region as channelized grids;
in step S3, the method for calculating the feature threshold for discrimination includes the steps of:
e1: presetting a natural number N as a maximum grid occupation frequency threshold;
constructing variables: grid occupancy time threshold;
traversing natural numbers within N from 1, and assigning the natural numbers to the grid occupation time threshold value one by one;
based on the assigned grid occupation times threshold, judging the occupation state of the space grids corresponding to each grid occupation times threshold, obtaining the road space topological structure corresponding to each grid occupation times threshold, and marking as: a topology for computation;
e2: and respectively taking the corresponding road space characteristic values and the road space characteristic distances in all the topological structures for calculation as dependent variables, taking the grid occupation frequency threshold value as independent variables, constructing corresponding relation curves, and respectively marking as: a road space characteristic value curve and a road space characteristic distance curve;
And e3: finding the longest horizontal segment on the road space characteristic value curve, and finding the highest horizontal segment on the road space characteristic distance curve on the threshold change interval of the grid occupation frequency threshold corresponding to the longest horizontal segment, and marking the highest horizontal segment as: calculating a characteristic interval;
and e4: and acquiring threshold change intervals of the grid occupation times threshold corresponding to all the characteristic interval intervals for calculation, and finding the grid occupation times threshold with the minimum value as the characteristic threshold for discrimination.
The invention provides a plane road vehicle passing space identification method, which is characterized in that the range of a road space to be identified is obtained based on vehicle track data acquired by road front end sensing equipment, and the road space to be identified is divided into a plurality of grids conforming to conditions; then, the road space topological structure, the vehicle passing space boundary, the vehicle passable space and the channelized space are identified by determining the occupied state of the grid, and then the category of the road space is judged according to the road space topological structure, and the area of the vehicle passing space is calculated. The whole identification, classification and quantification process does not need to manually investigate and set parameters of the road space in advance, does not need other additional information such as a map and the like, is driven by vehicle track data, and ensures the objectivity and consistency of the identification result of the vehicle passing space; meanwhile, the whole method has smaller requirements on data quantity and calculation capability, does not need manual calibration and machine training, does not depend on a map, has accurate result and high calculation speed, and can monitor the utilization state and change condition of the vehicle passing space on the plane road in real time.
Drawings
FIG. 1 is a flow chart of a planar road vehicle traffic space recognition method of the present invention;
FIG. 2 is a satellite diagram of a roadway space to be identified in an embodiment;
FIG. 3 is an example of a road space to be identified according to vehicle trajectory point distribution rasterization in an embodiment;
FIG. 4 is an example of occupancy status of a grid in a road space to be identified when a discrimination feature threshold is taken to be 1;
FIG. 5 is an embodiment of a map of road space topology and grid occupancy time threshold correspondence;
FIG. 6 is an example of occupancy status of a grid in a road space to be identified when the discrimination feature threshold is taken to 10;
FIG. 7 is an example of a vehicle traffic space boundary and traffic grid for a roadway space to be identified;
FIG. 8 is an example of a vehicle-navigable space for a roadway space to be identified;
FIG. 9 is an example of a vehicle-navigable space assuming a central green belt exists at an intersection south entrance to a roadway space to be identified;
FIG. 10 is an example of a channeling space assuming a central green belt exists at the intersection south access road of the roadway space to be identified;
fig. 11 is an example of a vehicle passing space calculated based on the present method.
Detailed Description
As shown in fig. 1, the present application includes a method for identifying a traffic space of a planar road, which includes the following steps.
S1: acquiring vehicle track data to be analyzed according to a preset data acquisition time period and road front end sensing equipment in a specified range;
as shown in fig. 2, in the embodiment, the road space to be identified is an X-shaped intersection, the used vehicle track data is collected by millimeter wave radars arranged on signal machine bars in four directions of the intersection, the collection time period is 56 minutes to 27 minutes from 16 hours of 2022 8 months 9 days, and 30MB of vehicle track data is collected. The specific acquisition time period is determined according to the traffic flow on the road or other road characteristics. If the traffic flow of the road space to be identified is smaller, the acquisition time period can be prolonged, and otherwise, the acquisition time period can be shortened.
S2: based on the coverage areas of all vehicle track points in the vehicle track data to be analyzed, confirming the range of the road space to be identified, uniformly dividing the road space to be identified into grids, and marking as: a space grid.
In this embodiment, the original vehicle position coordinates of the vehicle track points acquired based on the vehicle track data are longitude and latitude coordinates in the WGS-84 coordinate system, and the longitude and latitude coordinates can be converted into plane projection coordinates by using a coordinate conversion function transform_from_crs in phton. The range of the road space to be identified can be determined by determining the boundary of the road space to be identified, and in order to ensure that the road space to be identified can be divided into a plurality of square grids with equal size, the boundary of the road space to be identified is determined according to the following formula:
The left boundary of the range of the road space to be identified is
The right boundary of the range of the road space to be identified is
The upper boundary of the range of the road space to be identified is
The lower boundary of the range of the road space to be identified is
Wherein x is max X is the maximum value of the abscissa in the vehicle position coordinates min Is the minimum value of the abscissa in the vehicle position coordinates, y max Y is the maximum value of the ordinate in the vehicle position coordinates min L is the side length of the space grid, which is the minimum on the ordinate in the vehicle position coordinates. As shown in fig. 3, according to the distribution of the vehicle track points in the embodiment, the road space to be identified is divided into a space composed of 113×116 grids. Because the vehicle track point is the projection center point of the vehicle on the road, the too small value of the side length L of the space grid can not accurately identify the occupation of the physical size of the vehicle on the road space, and the too large value can reduce the sensitivity of identifying the driving behaviors such as turning, lane changing and the like. In this embodiment, the grid side length is 3 meters with reference to the width of the domestic standard car and the width of the standard urban road lane.
S3: judging the occupation state of each space grid based on the characteristic threshold value for judgment, the coordinate relation of the vehicle track points and the space grids;
The occupancy state of the space grid includes: occupied and unoccupied.
Normally, since vehicle track points are left when a vehicle passes through the space grids, if a vehicle track point exists in a certain space grid, the space grid can be considered to be used for vehicles to pass, therefore, the vehicle passing space can be identified by judging whether the vehicle track point exists in the space grid, and the occupied state of the space grid with the vehicle track point is marked as occupied, otherwise, the space grid with the vehicle track point exists is marked as unoccupied. However, because the existing vehicle track tracking technology has limited accuracy and the running behavior of the vehicle has uncertainty, whether the space grid can be used for passing the vehicle can not be accurately judged only by whether the vehicle track points exist in the space grid, and therefore, the method judges by checking the accumulated number of the vehicle track points in the space grid within a certain time, and the possibility of false recognition is reduced. Therefore, in the concrete calculation, a threshold value is set for the accumulated number of various vehicle track points in space according to the concrete characteristics of the road space: and (3) a grid occupation time threshold, namely when the accumulated number of vehicle track points in the space grid exceeds the grid occupation time threshold, recording the occupation state as occupied, and otherwise recording the occupation state as unoccupied.
In practical application, a plurality of grid occupation time thresholds may exist in the same plane road vehicle traffic space, which can be used for accurately judging the occupation state of the space grid, in the method, in order to ensure the calculation precision and keep the vehicle track information at the same time, the grid occupation time threshold with the minimum value is used for carrying out subsequent calculation, and the grid occupation time threshold with the minimum value is recorded as: and a feature threshold for discrimination.
The specific method for determining the occupied state of a certain space grid is that vehicle track points belonging to the space grid in vehicle track data are screened based on the boundary of the space grid and vehicle track point coordinates, and when the number of accumulated vehicle track points in the space grid is not less than a characteristic threshold for discrimination, the occupied state of the space grid is judged to be occupied; the method specifically comprises the following steps:
b1: acquiring space grid coordinates to be judged;
the coordinates of the space grid to be judged are set as follows: left boundary x left The right boundary is x right The upper boundary is y up The lower boundary is y down ；
Initializing a counter for each grid to be determined: cg=0;
b2: vehicle track points in the vehicle track data to be analyzed are acquired one by one and recorded as: vehicle track point to be calculated (x veh ，y veh )；
Step b3 is implemented on each vehicle track point to be calculated until all the vehicle track points are compared with the space grid to be judged, and step b4 is implemented;
b3: comparing the coordinates of the vehicle track points to be calculated with the coordinates of the space grids to be judged;
if x left ≤x veh <x right And y is left ≤y veh <y right The vehicle track points to be calculated belong to the space grids to be judged, and Cg=Cg+1;
b4: comparing the obtained Cg with a characteristic threshold value for discrimination;
when Cg is equal to or greater than the characteristic threshold for discrimination, the occupancy state of the grid to be discriminated is set to: is occupied;
otherwise, set to unoccupied.
Fig. 4 shows the occupancy state of the grids in the road space to be identified when the discrimination feature threshold is taken to be 1, the gray grids represent the grids whose occupancy state is occupied.
S3: identifying a road space topological structure in the road space to be identified based on the occupation states of all the space grids in the road space to be identified; the road space topology is represented based on the road space eigenvalue, road space eigenvalue spacing.
The method for identifying the road space topological structure comprises the following steps:
a1: the method comprises the steps that a feature frame is formed by the outermost space grids of a road space to be identified, the space grids in the feature frame are traversed clockwise from any space grid in a non-occupied state in the feature frame, the total number of times of occurrence of the space grids in the occupied state is recorded as a feature value of the feature frame, and the feature value is recorded as: a base feature frame feature value;
The space grid in the occupied state is recorded for 1 time when continuously appearing;
if the space grid in the unoccupied state does not exist in the feature frame, marking the feature value of the basic feature frame as 1;
a2: taking the space grids which are directly connected with the space grids forming the feature frame in the road space surrounded by the feature frame to form a new feature frame, and repeating the step a1 until the space surrounded by the feature frame does not contain any space grids, namely, after all the space grids participate in calculation, obtaining a group of sequences consisting of the feature values of the basic feature frame, and marking the sequences as: a base feature frame feature value sequence;
a3: in the basic feature frame feature value sequence, the basic feature frame feature value with the highest occurrence frequency is found and recorded as: calculating the maximum number of continuous occurrence times of the road space characteristic value, and marking the maximum number as: road space feature spacing.
In the method, the recognition accuracy of the occupied state of the space grid is controlled through the characteristic threshold value for discrimination; the abnormal occupation of the space grids by the vehicles generated by non-passing actions such as temporary parking cannot be distinguished due to the fact that the characteristic threshold value for distinguishing is too small, so that the number of the space grids in the occupied state is larger than the actual number; judging that the occupation of the space grids by vehicles at the position with smaller traffic flow is neglected when the value of the characteristic threshold is too large, so that the number of the identified space grids in the occupied state is smaller than the actual number; therefore, by adjusting the characteristic threshold value for discrimination, not only the space grids in the unoccupied state can be ensured not to be identified as the occupied state, but also all occupied space grids can be ensured to be covered, thereby ensuring the accuracy of identifying the final vehicle passing space.
In the application, the feature threshold for distinguishing is found through the corresponding relation diagram of the road space topological structure and the grid occupation frequency threshold, and the method specifically comprises the following steps:
e1: presetting a natural number N as a maximum grid occupation frequency threshold;
constructing variables: grid occupancy time threshold;
traversing natural numbers within N from 1, and assigning the natural numbers to the grid occupation times threshold value one by one;
based on the assigned grid occupation times threshold, steps b 1-b 4 are implemented by taking the grid occupation times threshold as a characteristic threshold for discrimination, and comparing with Cg, so as to judge the occupation state of the space grid corresponding to each grid occupation times threshold, and further, steps a 1-a 3 are implemented based on the occupation state of the space grid, and the road space topological structure corresponding to each grid occupation times threshold is obtained through calculation and is recorded as: a topology for computation;
e2: the corresponding road space characteristic values and road space characteristic distances in all the topological structures for calculation are respectively used as dependent variable (dependent variable), the grid occupation frequency threshold value is used as independent variable (independent variable), a corresponding relation curve is constructed, and the corresponding relation curve is respectively recorded as: a road space characteristic value curve and a road space characteristic distance curve;
And e3: finding the longest horizontal segment on the road space characteristic value curve, and finding the highest horizontal segment on the road space characteristic distance curve on the threshold change interval of the grid occupation frequency threshold corresponding to the longest horizontal segment, and marking as: calculating a characteristic interval;
and e4: and acquiring threshold change intervals of grid occupation times thresholds corresponding to all the characteristic interval intervals for calculation, and finding the grid occupation times threshold with the smallest value as the characteristic threshold for discrimination.
The corresponding relation diagram of the road space topological structure and the grid occupation frequency threshold value can intuitively and clearly reflect the condition that the topological structure of the road space to be identified changes along with the change of the grid occupation frequency threshold value, and particularly as shown in fig. 5, in order to make the image more intuitive and easy to use, a road space characteristic value curve and a road space characteristic distance curve are simultaneously drawn in the same diagram, the abscissa is the grid occupation frequency threshold value, the left main ordinate is the road space characteristic value, the right secondary ordinate is the road space characteristic distance, the dotted line is the road space characteristic value curve, and the solid line is the road space characteristic distance curve.
In the embodiment of fig. 5, the ordinate of the longest horizontal segment in the curve of the road space feature value is 4, that is, the road space feature value with the highest occurrence frequency is 4, and the ordinate of the highest horizontal segment on the curve of the road space feature pitch corresponding to the longest horizontal segment is 20, that is, the maximum road space feature pitch continuously occurring in the road space topology with the road space feature value of 4 is 20, and only one threshold interval corresponding to the maximum road space feature pitch continuously occurring is [10,13], so that the feature threshold value is taken as 10 for the discrimination.
FIG. 6 illustrates the occupancy state of the grid in the road space to be identified when the discriminating characteristic threshold is 10 in the embodiment; gray grids represent grids whose occupied states are occupied; when the characteristic threshold value for discrimination is 10, the characteristic value sequence of the basic characteristic frame obtained in the process of identifying the road space topological structure is [0,6,4,4,5,4,4,4,4,4,5,5,5,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,5,4,4,5,4,4,4,4,3,4,3,4,6,5,6,5,3,1,1,1,1,1,1], wherein the characteristic value of the basic characteristic frame with highest occurrence frequency is 4, namely the road space characteristic value; the sub-sequence of the basic feature frame feature values consisting of the continuously occurring road space feature values is [4,4], [4,4,4,4,4], [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4], [4,4], [4,4,4,4] in sequence, so that the maximum number of continuously occurring road space feature values is 20, namely the road space feature distance.
S5: and finding all the sub-sequences of the characteristic values of the basic characteristic frames formed by the characteristic values of the road space in the characteristic value sequence of the basic characteristic frames, wherein the characteristic frame corresponding to the characteristic value of the first basic characteristic frame of the sub-sequence with the longest length is the boundary of the vehicle passing space in the road space to be identified.
If the vehicle passing space boundary consists of the outermost space grid of the road space to be identified, selecting a feature frame corresponding to the feature value of the second basic feature frame of the subsequence with the longest length as the vehicle passing space boundary;
the space grid in the occupied state in the boundary of the vehicle passing space is marked as: a pass grid; and marking space grids except the passing grid in the road space to be identified as: a non-passing grid.
In the process of identifying the road space topological structure, the embodiment shown in fig. 6 determines the road space feature distance, and in the reference sub-sequence of the feature values of the basic feature frame composed of the road space feature values, the feature frame corresponding to the first feature value of the sub-sequence with the longest length is shown in fig. 7, and the black grid represents the vehicle passing space boundary of the road space to be identified in the embodiment. Meanwhile, in fig. 7, gray grids represent passing grids, and white grids represent non-passing grids.
In practice, there may be some space grid that is identified as being in an opposite state to its desired occupancy state during the occupancy state identification process. Such as: in a non-vehicle passing space beside a road, since a vehicle is parked for a long time in a certain space grid, the occupied state of the space grid may be erroneously recognized as: is occupied; also, in a traffic road space, because a certain space grid has not passed through a vehicle due to road maintenance or the like, the occupied state of the space grid may be erroneously recognized as: and is unoccupied. Misrecognition of the space grid occupancy state may result in misrecognition of the passing and non-passing grids, such as: the pass-through grids are present in isolation in one piece of non-pass-through grids, or the non-pass-through grids are present in isolation in one piece of pass-through grids. In addition, traffic grids at the edges of road spaces may be subject to discontinuities due to limitations in the quality or quantity of vehicle trajectory data to be analyzed.
S6: screening and re-marking the passing grid and the non-passing grid; according to the connected domain calculation method, the connected domain formed by the passing grids is a vehicle passable space in the road space to be identified;
the occupancy state of the space grid in the vehicle-passable space is: is occupied;
the occupation state of the space grids which do not belong to the space where vehicles can pass in the road space to be identified is as follows: and is unoccupied.
The specific steps of screening and re-marking the passing grid and the non-passing grid include:
c1: extracting a connected domain formed by passing grids, and marking the connected domain as: passing through the grid connected domain;
the passing grids which do not belong to the largest passing grid connected domain are re-marked as non-passing grids, namely: finding a pass grid existing in isolation in a piece of non-pass grid, and marking the pass grid as the non-pass grid;
the largest passing grid connected domain comprises the connected domain with the largest passing grid quantity;
c2: extracting a connected domain consisting of non-passing grids, and marking the connected domain as: a non-passing grid connected domain;
re-labeling the non-passing grid which does not belong to the largest non-passing grid connected domain as a passing grid, namely: finding a non-passing grid existing in isolation in a piece of passing grid, and marking the non-passing grid as the passing grid;
Wherein the largest non-passing grid connected domain comprises the connected domain with the largest non-passing grid number;
c3: checking the arrangement condition of the passing grids and the non-passing grids row by row and column by column, and if two passing grids are separated by one non-passing grid, re-marking the non-passing grid as the passing grid, namely: finding a condition that the arrangement of the passing grids at the edge of the road space is discontinuous, and marking the passing grids as passing grids;
c4: the occupancy state of the passing grid is re-marked as: is occupied;
the occupancy state of the non-passing grid is re-marked as: and is unoccupied.
The connected domain extraction method comprises the following steps:
based on a four-communication marking algorithm or an eight-communication marking algorithm, each kind of space grid (such as a passing grid) is belonged to one kind of communication domain (such as a passing grid communication domain). In the method, the connected domain extraction method is described by taking the passing grids as an example, and attributing the passing grids, the non-passing grids and the subsequent channeling grids to the connected domains of the corresponding types.
f1: any passing grid is taken and is recorded as a current connected domain;
f2: taking any passing grid which does not belong to any connected domain and does not participate in extraction of the current connected domain, and recording the passing grid as a grid to be judged;
f3: judging whether the grid to be judged and any grid in the current communication domain have public characteristics or not;
when extracted based on a four-way token algorithm, the common features are: public edges, when extracted based on eight-connectivity-labeling algorithm, are characterized by: a common vertex;
if there are common features, then step f4 is performed;
otherwise, implementing step f5;
f4: adding the grid to be judged into the current connected domain;
f5: marking the grid to be judged as a reckoning grid;
f 2-f 5 are circularly implemented until all the passing grids participate in the extraction of the current connected domain, namely the passing grids belong to the current connected domain or are marked as reckoning grids, and then the step f6 is implemented;
f6: f 1-f 6 are circularly implemented based on the reckoning grids until all the passing grids belong to a certain connected domain, and the algorithm is finished.
And according to the actual data quality of the vehicle track data to be analyzed and the specific application requirements of vehicle traffic space identification, a four-connected marking algorithm or an eight-connected marking algorithm is properly selected to extract the connected domain.
Fig. 8 shows the vehicle-accessible space extracted after screening and re-marking the pass and non-pass grids of fig. 7, the space formed by gray grids representing the vehicle-accessible space of the road space to be identified in the embodiment.
In actual road spaces, a green belt type of separator is often provided, and although vehicles cannot pass through the separator, in traffic organization and management, the green belt type of separator is generally used as a reserve space for a vehicle passing space, and therefore, the present method regards the space occupied by such a separator as a part of the vehicle passing space, and records as a trenched space. Since the space grid in the channeling space is clearly unoccupied, it is identified as a non-passing grid in steps S1-S5. Because the data collection range of the road front-end sensing device is limited, the non-passing grid corresponding to the channeling space is not completely surrounded by the passing grid, so in step S6, the non-passing grid corresponding to the channeling space is not re-marked as the passing grid. In order to identify the channeling space in the vehicle passing space, the method extracts the channeling grids in the vehicle passing space boundary and the communicating domain formed by the channeling grids as the channeling space according to the identified vehicle passing space boundary and the screened passing grids and non-passing grids.
S7: based on the passing grids and the non-passing grids, extracting channelized grids in a vehicle passing space, and marking a connected domain formed by the channelized grids as: a channeling space;
The occupancy state of the space grid within the channeling space is re-labeled as: is occupied.
A method of extracting a canalized grid in a vehicle passing space, comprising the steps of:
d1: starting from any one of the grids forming the vehicle passing space boundary, traversing the grids in the vehicle passing space boundary sequentially in the same direction, recording the number of non-passing grids when continuously appearing, and forming: a collection of non-passing grid numbers;
d2: adding elements in the non-passing grid quantity set: 1, a step of;
d3: dividing the number set in the step d2 into two types through a clustering algorithm;
if 1 is singly classified into one type, the canalization grid does not exist, and the algorithm is ended;
otherwise, implementing step d4;
d4: continuously occurring non-passing grids corresponding to the number 1 belonging to the same class are re-labeled: a trenching grid;
d5: and extracting a connected domain formed by non-passing grids in the road space to be identified, and re-marking the non-passing grids which do not belong to the largest non-passing grid connected domain as channelized grids.
In the embodiment shown in fig. 8, the grids in the vehicle passing space boundary are traversed anticlockwise from the leftmost passing grid on the lower boundary of the vehicle passing space boundary of the road space to be identified, the passing grids are marked as 1, the non-passing grids are marked as 0, and the grid occupation record [1,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0] is obtained, so that the number set of non-passing grids which continuously appear is {81,68,95,68}; the k-means clustering algorithm is adopted to classify elements in the set {81,68,95,68,1}, and the elements in the set can be classified into two types {81,68,95,68} and {1 }. Thus, the road space to be identified in an embodiment is free of channeling grids and channeling spaces.
Assuming that the central green belt exists at the intersection south entrance of the road space to be identified in the embodiment, since the green belt cannot travel the vehicle, it is identified as a non-passing grid in steps S1 to S6, as shown in fig. 9.
The grid occupation record of the vehicle passing space boundary is [1,1,1,1,1,0,0,0,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,1,1,1,1,1,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0], the number set of non-passing grids appearing continuously is {3,81,68,95,68}, the set {3,81,68,95,68,1} is classified by adopting a k-means clustering algorithm, and elements in the set are classified into {81,68,95,68} and {3,1 }. The grid corresponding to the element {3}, i.e. the 6 th to 8 th grids in the grid occupation record, is re-marked as a channelized space extracted after the channelized grid is shown in fig. 10. The space formed by the black grids represents the canal space when the central green belt exists at the south entrance road of the intersection of the road space to be identified in the embodiment.
In the method, the occupied state of the channeling grid is recorded as occupied because the channeling space can be converted into a vehicle-passable space in traffic organization and management.
S8: and identifying the road space topological structure in the road space to be identified based on the latest occupied states of all the space grids in the road space to be identified.
In the embodiment shown in fig. 8, the road space to be identified does not have a canalization grid and a canalization space, and the vehicle passing space in the embodiment is the vehicle passable space, as shown in fig. 11, the gray grid represents the vehicle passing space identified in the embodiment; even if the road space to be identified in the embodiment has the canalization grid and the canalization space, the vehicle passing space can be correctly identified as shown in fig. 11; the road space topology of the vehicle passing space can still be correctly identified as [4,20], i.e. the road space feature value is 4 and the road space feature distance is 20.
S9: the specific method for judging the traffic space category of the vehicle according to the road space characteristic value comprises the following steps:
when the road space characteristic value is not more than 2, the vehicle passing space class is a road section;
when the road space characteristic value is equal to 3, the vehicle passing space class is T-shaped or Y-shaped intersection;
When the road space characteristic value is equal to 4, the vehicle passing space class is a cross or X-shaped road opening;
and when the road space characteristic value is greater than 4, the vehicle passing space class is a multi-entrance road junction.
In the embodiment shown in fig. 11, by identifying the road space topology of the vehicle passing space, since the road space feature value of the vehicle passing space is 4, the vehicle passing space category in the embodiment is judged to be a cross or an X-shaped intersection.
S10: the area of the vehicle passing space is calculated by the following method:
S veh ＝S cell (N veh +N cana )
wherein S is veh Is the area of the vehicle passing space S cell Is the area of the space grid, N veh N being the number of space grids comprised in the vehicle-operable space cana The number of space grids included in the trenched space.
In the embodiment shown in fig. 11, the area of the space grid is 9 square meters, the number of space grids included in the vehicle-passable space is 1786, and the number of space grids included in the channeling space is 0, so that the vehicle-passing space area is 16074 square meters.
The planar road vehicle passing space identification method provided by the invention can accurately identify the topological structure of the road space, the type and the area of the road space, has the advantages of small data demand, no need of manual calibration and machine training, no dependence on maps and the like, and can be applied to the scenes of real-time monitoring of the utilization state of the road space, the research and judgment analysis of road congestion problems, the auxiliary decision of traffic organization measures, the comparison and optimization of signal timing schemes and the like. The technical scheme of the invention can energize the front-end sensing equipment, acquire a small amount of data to quickly update the topology structure, the category, the area and other information of the road traffic space, and continuously acquire the data in real time on the basis, so as to realize real-time monitoring of the road space utilization state; if the technical scheme of the invention is operated for a long time, the front-end sensing equipment accumulates the road space utilization data, is helpful for summarizing the utilization rule of vehicles to road traffic space, and supports the automatic analysis, the study and judgment of the road congestion problem and the auxiliary decision of traffic organization measures; the technical scheme of the invention can also support the comparison and optimization of the signal timing schemes by simulating the utilization state of the road space of the crossing under different signal timing schemes under the condition that the front-end sensing equipment is communicated with the crossing signal machine system. In conclusion, the technical scheme of the invention provides a set of planar road vehicle passing space identification, classification and quantification method which is low in cost, high in performance and easy to popularize, and has wide application prospect in the development of directly enabling the existing road front end sensing equipment to promote traffic refinement and intelligent treatment.
Claims (7)
1. A method for identifying a traffic space of a planar road, comprising the steps of:
s1: acquiring vehicle track data to be analyzed according to a preset data acquisition time period and road front end sensing equipment in a specified range;
the method is characterized by further comprising the following steps:
s2: based on the coverage areas of all vehicle track points in the vehicle track data to be analyzed, confirming the space range of the road space to be identified, uniformly dividing the road space to be identified into grids, and marking as: a space grid;
s3: judging the occupation state of each space grid based on the characteristic threshold value for judgment, the coordinate relation of the vehicle track point and the space grid;
the occupancy state of the space grid comprises: occupied and unoccupied;
s4: identifying a road space topological structure in the road space to be identified based on the occupation states of all the space grids in the road space to be identified; the road space topological structure is expressed based on [ road space characteristic values, road space characteristic intervals ];
the method for identifying the road space topological structure comprises the following steps:
a1: the space grids at the outermost side of the road space to be identified form a feature frame, the space grids in the feature frame are traversed clockwise from any space grid in a non-occupied state in the feature frame, the total number of times of occurrence of the space grids in the occupied state is recorded as a feature value of the feature frame, and the feature value is recorded as: a base feature frame feature value;
Recording only 1 time when the space grid in the occupied state continuously appears;
if the space grid in the unoccupied state does not exist in the feature frame, marking the feature value of the basic feature frame as 1;
a2: taking the space grids which are directly connected with the space grids forming the feature frame in the road space surrounded by the feature frame, forming a new feature frame, and repeating the step a1 until the space surrounded by the feature frame does not contain any space grids, namely, after all the space grids participate in calculation, obtaining a group of sequences consisting of the feature values of the basic feature frame, and marking the sequences as: a base feature frame feature value sequence;
a3: in the basic feature frame feature value sequence, finding the basic feature frame feature value with highest occurrence frequency is recorded as: calculating the maximum number of continuous occurrence times of the road space characteristic value, and recording the maximum number as: road space feature spacing;
s5: finding all the subsequences of the characteristic values of the basic characteristic frames formed by the characteristic values of the road space in the characteristic value sequence of the basic characteristic frames, wherein the characteristic frame corresponding to the characteristic value of the first basic characteristic frame of the subsequence with the longest length is the traffic space boundary of the vehicle in the road space to be identified;
If the vehicle passing space boundary consists of the space grid at the outermost side of the road space to be identified, selecting the feature frame corresponding to the feature value of the second basic feature frame of the sub-sequence with the longest length as the vehicle passing space boundary;
the space grid in the occupied state in the boundary of the vehicle passing space is marked as: a pass grid; and marking the space grids except the passing grid in the road space to be identified as: a non-passing grid;
s6: screening and re-marking the pass grid and the non-pass grid; according to a connected domain calculation method, the obtained connected domain formed by the passing grids is a vehicle passable space in the road space to be identified;
the occupancy state of the space grid in the vehicle-navigable space is: is occupied;
the occupation state of the space grid which does not belong to the vehicle-passable space in the road space to be identified is as follows: non-occupation;
s7: extracting a channelized grid in the vehicle passing space based on the passing grid and the non-passing grid, wherein a connected domain formed by the channelized grid is marked as: a channeling space;
Re-labeling the occupancy state of the space grid within the canalized space as: is occupied;
s8: identifying the road space topology structure in the road space to be identified based on the latest occupied states of all the space grids in the road space to be identified;
s9: the vehicle passing space category is judged according to the road space characteristic value, and the specific method comprises the following steps:
when the road space characteristic value is not more than 2, the vehicle passing space class is a road section;
when the road space characteristic value is equal to 3, the vehicle passing space class is a T-shaped or Y-shaped intersection;
when the road space characteristic value is equal to 4, the vehicle passing space class is a cross or X-shaped road opening;
when the road space characteristic value is greater than 4, the vehicle passing space class is a multi-entrance road junction;
s10: the area of the vehicle passing space is calculated by the following method:
S veh ＝S cell (N veh +N cana )
wherein S is veh For the area of the vehicle passing space S cell For the area of the space grid, N veh For the number, N, of the space grids contained in the vehicle-operable space cana And (3) the number of the space grids contained in the canalized space.
2. A method for identifying a traffic space of a planar road according to claim 1, wherein: the calculation method of the range of the road space to be identified comprises the following steps:
The left boundary of the range of the road space to be identified is
The right boundary of the range of the road space to be identified is
The upper boundary of the range of the road space to be identified is
The lower boundary of the range of the road space to be identified is
Wherein x is max X is the maximum value of the abscissa in the vehicle position coordinates min Is the minimum value of the abscissa in the vehicle position coordinates, y max Y is the maximum value of the ordinate in the vehicle position coordinates min And L is the minimum value of the ordinate in the vehicle position coordinates, and L is the side length of the space grid.
3. A method for identifying a traffic space of a planar road according to claim 1, wherein: the method for judging the occupancy state of the space grid comprises the following steps:
b1: acquiring space grid coordinates to be judged;
the coordinates of the space grid to be judged are set as follows: left boundary x left The right boundary is x right The upper boundary is y up The lower boundary is y down ；
Initializing a counter for each of the grids to be judged: cg=0;
b2: the vehicle track points in the vehicle track data to be analyzed are acquired one by one and recorded as: vehicle track point to be calculated (x veh ，y veh )；
B3 is implemented on each vehicle track point to be calculated until all the vehicle track points are compared with the space grid to be judged, and step b4 is implemented;
b3: comparing the coordinates of the vehicle track points to be calculated with the coordinates of the space grid to be judged;
if x left ≤x veh <x right And y is left ≤y veh <y right The vehicle track points to be calculated belong to the space grid to be judged, and cg=cg+1;
b4: comparing the obtained Cg with the characteristic threshold for discrimination;
when Cg is greater than or equal to the characteristic threshold for discrimination, the occupancy state of the grid to be discriminated is set as: is occupied;
otherwise, set to unoccupied.
4. A method for identifying a traffic space of a planar road according to claim 1, wherein: in step S6, the specific steps of screening and re-marking the passing grid and the non-passing grid include:
c1: extracting a connected domain formed by the passing grids, and marking the connected domain as: passing through the grid connected domain;
re-marking the passing grid which does not belong to the largest passing grid connected domain as a non-passing grid;
c2: extracting a connected domain consisting of the non-passing grids, and marking the connected domain as: a non-passing grid connected domain;
re-marking the non-passing grid which does not belong to the largest non-passing grid connected domain as a passing grid;
c3: checking the arrangement condition of the passing grids and the non-passing grids row by row and column by column, and re-marking the non-passing grids as passing grids if the two passing grids are separated by one non-passing grid;
c4: the occupancy state of the passing grid is re-marked as: is occupied;
re-marking the occupancy state of the non-passing grid as: and is unoccupied.
5. A method for identifying a traffic space of a planar road according to claim 1, wherein: the connected domain extraction method comprises the following steps:
each kind of grid is attributed to one kind of connected domain based on a four-connected marking algorithm or an eight-connected marking algorithm.
6. A method for identifying a traffic space of a planar road according to claim 1, wherein: a method of extracting the canalized grid in the vehicle passing space, comprising the steps of:
d1: starting from any one of the grids forming the vehicle passing space boundary, traversing the grids in the vehicle passing space boundary sequentially in the same direction, and recording the number of the non-passing grids when continuously appearing to form: a collection of non-passing grid numbers;
d2: adding elements in the non-passing grid quantity set: 1, a step of;
d3: dividing the number set in the step d2 into two types through a clustering algorithm;
if 1 is singly classified into one type, the canalization grid does not exist, and the algorithm is ended;
Otherwise, implementing step d4;
d4: continuously occurring non-passing grids corresponding to the number 1 belonging to the same class are re-labeled: a trenching grid;
d5: and extracting a connected domain formed by the non-passing grids in the road space to be identified, and re-marking the non-passing grids which do not belong to the largest non-passing grid connected domain as channelized grids.
7. A method for identifying a traffic space of a planar road according to claim 1, wherein: in step S3, the method for calculating the feature threshold for discrimination includes the steps of:
e1: presetting a natural number N as a maximum grid occupation frequency threshold;
constructing variables: grid occupancy time threshold;
traversing natural numbers within N from 1, and assigning the natural numbers to the grid occupation time threshold value one by one;
based on the assigned grid occupation times threshold, judging the occupation state of the space grids corresponding to each grid occupation times threshold, obtaining the road space topological structure corresponding to each grid occupation times threshold, and marking as: a topology for computation;
e2: and respectively taking the corresponding road space characteristic values and the road space characteristic distances in all the topological structures for calculation as dependent variables, taking the grid occupation frequency threshold value as independent variables, constructing corresponding relation curves, and respectively marking as: a road space characteristic value curve and a road space characteristic distance curve;
And e3: finding the longest horizontal segment on the road space characteristic value curve, and finding the highest horizontal segment on the road space characteristic distance curve on the threshold change interval of the grid occupation frequency threshold corresponding to the longest horizontal segment, and marking the highest horizontal segment as: calculating a characteristic interval;
and e4: and acquiring threshold change intervals of the grid occupation times threshold corresponding to all the characteristic interval intervals for calculation, and finding the grid occupation times threshold with the minimum value as the characteristic threshold for discrimination.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310154797.XA CN116168538B (en) | 2023-02-23 | 2023-02-23 | Planar road vehicle passing space identification method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310154797.XA CN116168538B (en) | 2023-02-23 | 2023-02-23 | Planar road vehicle passing space identification method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116168538A true CN116168538A (en) | 2023-05-26 |
| CN116168538B CN116168538B (en) | 2024-04-12 |
Family
ID=86414430
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310154797.XA Active CN116168538B (en) | 2023-02-23 | 2023-02-23 | Planar road vehicle passing space identification method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116168538B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117131149A (en) * | 2023-10-26 | 2023-11-28 | 四川国蓝中天环境科技集团有限公司 | Earth and rock point location and transportation network identification method based on GPS track of slag transport vehicle |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160102987A1 (en) * | 2014-10-14 | 2016-04-14 | Guangzhou Hkust Fok Ying Tung Research Institute | Method for inferring type of road segment |
| WO2021077761A1 (en) * | 2019-10-23 | 2021-04-29 | 江苏智通交通科技有限公司 | Intersection abnormal vehicle trajectory identification and analysis method based on hierarchical clustering |
| CN113505187A (en) * | 2021-07-07 | 2021-10-15 | 西安理工大学 | Vehicle classification track error correction method based on map matching |
| CN113570860A (en) * | 2021-07-26 | 2021-10-29 | 福州大学 | Method for finely dividing and identifying urban road traffic states aiming at sparse track data |
| CN113936481A (en) * | 2021-12-02 | 2022-01-14 | 公安部交通管理科学研究所 | Method for evaluating space utilization condition of signal control intersection |
-
2023
- 2023-02-23 CN CN202310154797.XA patent/CN116168538B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20160102987A1 (en) * | 2014-10-14 | 2016-04-14 | Guangzhou Hkust Fok Ying Tung Research Institute | Method for inferring type of road segment |
| WO2021077761A1 (en) * | 2019-10-23 | 2021-04-29 | 江苏智通交通科技有限公司 | Intersection abnormal vehicle trajectory identification and analysis method based on hierarchical clustering |
| CN113505187A (en) * | 2021-07-07 | 2021-10-15 | 西安理工大学 | Vehicle classification track error correction method based on map matching |
| CN113570860A (en) * | 2021-07-26 | 2021-10-29 | 福州大学 | Method for finely dividing and identifying urban road traffic states aiming at sparse track data |
| CN113936481A (en) * | 2021-12-02 | 2022-01-14 | 公安部交通管理科学研究所 | Method for evaluating space utilization condition of signal control intersection |
Non-Patent Citations (3)
| Title |
|---|
| JINJUN TANG 等: "Exploring dynamic property of traffic flow time series in multi-states based on complex networks: Phase space reconstruction versus visibility graph", PHYSICA A 450, 22 January 2016 (2016-01-22) * |
| 李翔;马爽;杨辉;张晓楠;: "矢量道路拓扑追踪匹配算法", 中国图象图形学报, no. 05, 16 May 2017 (2017-05-16) * |
| 葛宇然 等: "基于时空联合学习的城市交通流短时预测模型", 计算机工程, vol. 49, no. 1, 31 January 2023 (2023-01-31) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117131149A (en) * | 2023-10-26 | 2023-11-28 | 四川国蓝中天环境科技集团有限公司 | Earth and rock point location and transportation network identification method based on GPS track of slag transport vehicle |
| CN117131149B (en) * | 2023-10-26 | 2024-01-23 | 四川国蓝中天环境科技集团有限公司 | Earth and rock point location and transportation network identification method based on GPS track of slag transport vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116168538B (en) | 2024-04-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110570658B (en) | Method for identifying and analyzing abnormal vehicle track at intersection based on hierarchical clustering | |
| CN110728842B (en) | Abnormal driving early warning method based on reasonable driving range of vehicles at intersection | |
| CN108415975B (en) | BDCH-DBSCAN-based taxi passenger carrying hot spot identification method | |
| CN109254861B (en) | OD demand extraction and reliability analysis method based on track data | |
| CN108648445B (en) | Dynamic traffic situation prediction method based on traffic big data | |
| CN110929939B (en) | Landslide hazard susceptibility spatial prediction method based on clustering-information coupling model | |
| CN101807345A (en) | Traffic jam judging method based on video detection technology | |
| CN116168538B (en) | Planar road vehicle passing space identification method | |
| CN108961758A (en) | A kind of crossing broadening lane detection method promoting decision tree based on gradient | |
| CN108665703A (en) | Road network state transition point method of discrimination based on macroscopical parent map | |
| CN111882858B (en) | Multi-source data-based method for predicting queuing length of highway abnormal event | |
| CN113327248B (en) | Tunnel traffic flow statistical method based on video | |
| CN111860383B (en) | Group abnormal behavior identification method, device, equipment and storage medium | |
| CN108629982B (en) | Road section vehicle number estimation method based on travel time distribution rule | |
| CN110598999A (en) | Traffic travel analysis method, system and storage medium based on individual data | |
| CN111640304A (en) | Automatic quantitative extraction method for traffic jam propagation characteristics of continuous flow traffic facility | |
| CN112085951A (en) | Traffic state discrimination method, system, storage medium, computer device and application | |
| CN114049788B (en) | Parking blind spot identification method and device, electronic device and computer program product | |
| Chen et al. | Research on the influencing factors of elderly pedestrian traffic accidents considering the built environment | |
| CN116704750B (en) | Traffic state identification method based on clustering algorithm, electronic equipment and medium | |
| CN113936481B (en) | Method for evaluating space utilization condition of signal control intersection | |
| CN106446194A (en) | Determination method and device for acquisition parameters of a geographic region | |
| CN116628531A (en) | Crowd-sourced map road object element clustering method, system and storage medium | |
| CN112309129B (en) | OD matrix forming method for vehicle travel characteristic classification based on checkpoint license plate recognition | |
| CN115273468A (en) | Traffic congestion control strategy generation method and device |
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 |