CN109767615A - Road network traffic flow key flow direction and critical path analysis method - Google Patents
Road network traffic flow key flow direction and critical path analysis method Download PDFInfo
- Publication number
- CN109767615A CN109767615A CN201811227838.9A CN201811227838A CN109767615A CN 109767615 A CN109767615 A CN 109767615A CN 201811227838 A CN201811227838 A CN 201811227838A CN 109767615 A CN109767615 A CN 109767615A
- Authority
- CN
- China
- Prior art keywords
- track
- cluster
- trip
- road network
- critical path
- 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
Landscapes
- Traffic Control Systems (AREA)
Abstract
The present invention provides a kind of road network traffic flow key flow direction and critical path analysis method, the track of vehicle generated based on bayonet data, track fractionation is carried out using Spectral Clustering, the space-time characteristic variable of trip track is extracted on this basis, the cluster of trip track is realized by topic model, to identify the crucial flow direction and critical path of road network；Present invention employs to missing information have higher " tolerance " degree topic model track of vehicle is analyzed, and then from track of vehicle extract road network traffic flow key flow to and critical path, provide important support for road network Demand-side signature analysis.
Description
Technical field
The present invention relates to a kind of road network traffic flow key flow direction and critical path analysis methods.
Background technique
The large-scale application of various advanced communications and computer technology on road network, so that obtaining single unit vehicle on road
Online complete trip track is possibly realized.Since vehicle driving track includes trip information abundant, as travel time,
Trip route etc., thus not only had great advantages in terms of the trip characteristics of the single individual of analysis, it is whole for research road network
Body travel behaviour characteristic aspect, vehicle driving track also have very high real value and advantage.
Vehicle identification and vehicle location can achieve the effect that Vehicle tracing.It is obtained based on video bayonet data
Track of vehicle is exactly the typical case of vehicle identification.But it is different from through GPS device real-time vehicle tracing point obtained, one
Aspect is substantially the actual position of equipment based on track of vehicle acquired in video bayonet data, that is, passes through its institute
What is obtained is a kind of " sparse " track；On the other hand, it since video identification equipment has " missing inspection ", thus will lead to
The phenomenon that track lacks.
In view of the intelligent measurement condition of Current City Road, video card make a slip of the tongue car test measured data procurement cost it is lower,
It being capable of popularization and application.Given this how status provides a kind of road network traffic flow key flow direction and critical path analysis method, needle
The characteristics of to the tracks of vehicle of bayonet data, the extraction of Lai Shixian road network traffic flow key flow direction and critical path, is that should give
To consider the problems of and solve.
Summary of the invention
The object of the present invention is to provide a kind of road network traffic flow key flow directions and critical path analysis method to solve existing skill
Present in art on the one hand, based on track of vehicle acquired in video bayonet data, it is substantially the actual cloth set of equipment
It sets, i.e., obtained by its is a kind of " sparse " track；On the other hand, since video identification equipment has feelings such as " missing inspections "
Condition, thus will lead to track missing the phenomenon that the problem of.
The technical solution of the invention is as follows:
A kind of road network traffic flow key flow direction and critical path analysis method, based on the track of vehicle that bayonet data generate,
Track fractionation is carried out using Spectral Clustering, the space-time characteristic variable of trip track is extracted on this basis, passes through topic model
It realizes the cluster of trip track, to identify the crucial flow direction and critical path of road network, specifically includes following steps,
S1, the history for reading tollgate devices detection cross car data, according to the brand number crossed in car data, for daily
All vehicles for passing through bayonet position in road network generate the complete trip track on the vehicle same day；
S2, original trip track is split as by several sub-trajectory Tr={ tr based on spectral clustering1,···,tri,···,
trm}；
S3, the space-time characteristic Variables Sequence that trip track is extracted based on discrete Fourier transform DFT；
S4, the trip track space-time characteristic Variables Sequence in historical period is obtained based on step S3, using Di Li Cray mistake
Journey mixed model carries out clustering to trip track；
S5, the cluster result based on step S4, using the cluster centre of each cluster as the core trip track of corresponding cluster；
The two line angle is calculated according to the coordinate of trip track starting point to the end, thereby determines that the traffic flow key flow direction of road network；It adopts
With the tracing point in K shortest path algorithm processing core trip track, K critical path is determined.
Further, in step S1, the complete trip track on the vehicle same day is generated, specifically, the mistake chronologically arranged
Vehicle point set X={ x1,…,xi,…,xn}；Wherein xi=(pi,ti), piIt is true according to tollgate devices position for tracing point coordinate
It is fixed；tiTo be determined by the vehicle time that crosses corresponded in car data by the track point moment；N crossed car data for the vehicle same day
Quantity.
Further, it in step S1, crosses car data and includes device numbering, spends vehicle time, brand number.
Further, step S2 specifically,
S21, similar matrix S, the element in matrix are established using gaussian kernel function RBFWherein a, b ∈ [1, n], ρp、 ρtRespectively track
Point coordinate passes through the standard deviation of track point moment；
S22, adjacency matrix W=S, i.e., wherein element wa,b=sa,b；Degree matrix D is diagonal matrix,
S23, building Laplacian Matrix L=D-W, do standardization: D-1/2LD-1/2, most by the matrix after standardization
Feature vector corresponding to k small characteristic value forms n*k dimensional feature matrix F；
S24, the every a line composition 1*k for extracting F tie up sample matrix f, use DBSCASN clustering algorithm to whole n samples
Clustering is carried out, cluster division result { F is obtained1,···,Fi,···,Fm, wherein m is the number for the cluster that cluster generates
Amount, cluster FiIt is made of several sample matrix；According to the sample matrix in each cluster, the cluster division result of tracing point is determined
TR={ tr1,···,tri,···,trm, wherein element is single trip sub-trajectory, tri={ x(i,1),
x(i,2),···,x(i,u),···,x(i,h), wherein x(i,u)For sub-trajectory triInterior tracing point, h are the trip sub-trajectory
The tracing point quantity of approach.
Further, step S3 specifically,
S31, for any sub-trajectory tri={ x(i,1),x(i,2),···,x(i,u),···,x(i,h), extract track
The location dimension of point, i.e. P_tri={ p(i,1),p(i,2),···,p(i,u),···,pi,h), p(i,j)=[lng(i,u),
lat(i,u)], lng(i,u)、lat(i,u)Respectively position warp, latitude coordinate；
S32, the corresponding Fourier coefficient f_tr of the sub-trajectory is calculatedi=(fi x,fi y), wherein For
Imaginary part；
S33, the Fourier coefficient F=(f_tr for generating trip track1,f_tr2,···,f_trm)。
Further, in step S4, Di Li Cray process mixed model specifically:
Wherein, Dirichlet (*) is dirichlet function,For the probability of different clusters, β is that the control of clusters number is joined
Number, zlFor the l articles trip track TRlAffiliated cluster, T are trip tracking quantity to be clustered, θcFor the variance of c-th of cluster, ε is
Initial clustering number；α is the sub-trajectory number of satisfaction needed for a certain track is divided into certain cluster defined in model；xl,uFor
The l articles trip track TRlIn u-th of sub-trajectory；
Using the probability for folding Gibbs sampling method approximate solution Di Li Cray model, x and z joint probability point are obtained
Cloth function is shown below:
Wherein, P (zl=c | z-l,TR1:M, α, β) and it is the probability that the l articles trip track TRl belongs to c-th of cluster； z-lFor not
Include the track TR that goes on a journeylGathering close；TR1:MIndicate trip track bulk sample sheet；Nc(z-l) it is to reject trip track TRlC afterwards
It goes on a journey in a cluster tracking quantity；FlFor the track TR that goes on a journeylFourier coefficient； F_columnc(TR-l) it is to be rejected in c-th of cluster
Go on a journey track TRlThe column vector that the Fourier coefficient of other all trip tracks is constituted afterwards；Parameter in formulaWherein Γ (*) is gamma function, and τ is gamma function parameter；In order to maximize P (zl=c | z-l,
TR1:M, α, β), it is iterated sampling using Gibbs model, finally obtains the parameter Estimation of Di Li Cray process mixed model
Value.
The beneficial effects of the present invention are:
One, this kind of road network traffic flow key flow direction and critical path analysis method, for the track of vehicle of bayonet data
Feature provides a kind of road network entirety travel behaviour characteristic analysis method, wherein crucial flow direction is to representative in road network
The description in Macro-traffic Flow direction, and critical path is the maximum path of the vehicle select probability (collection by crucial flow direction traveling
It closes)；Present invention employs to missing information have higher " tolerance " degree topic model track of vehicle is analyzed, into
And road network traffic flow key flow direction and critical path are extracted from track of vehicle, it is provided for road network Demand-side signature analysis important
Support.
Two, the present invention is on the basis of the sub-trajectory that spectral clustering is handled, by discrete Fourier transform to the complete of vehicle
Whole trip track is handled, its space-time characteristic variable is extracted.
Three, for the shortcoming that can not automatically determine clusters number in spectral clustering, the present invention is by DBSCAN algorithm
It is embedded in traditional spectral clustering, realizes optimum clustering number purpose and automatically determine.
Four, the present invention is based on track space-time characteristics extracts the crucial stream of road network using Di Li Cray process mixed model
To and critical path.
Detailed description of the invention
Fig. 1 is the flow diagram of road network traffic flow key of embodiment of the present invention flow direction and critical path analysis method.
Specific embodiment
The preferred embodiment that the invention will now be described in detail with reference to the accompanying drawings.
Embodiment
A kind of road network traffic flow key flow direction and critical path analysis method, based on the track of vehicle that bayonet data generate,
Track fractionation is carried out using Spectral Clustering, the space-time characteristic variable of trip track is extracted on this basis, passes through topic model
It realizes the cluster of trip track, to identify the crucial flow direction and critical path of road network, specifically includes following steps,
S1, read tollgate devices detection history cross car data, wherein cross car data include device numbering, cross the vehicle time,
Brand number；Reject bus, taxi crosses vehicle record；According to the brand number crossed in car data, for institute in daily road network
There is the vehicle for passing through bayonet position, generate the complete trip track on the vehicle same day, i.e., what is chronologically arranged crosses vehicle point
Set X={ x1,…,xi,…,xn}；Wherein xi=(pi,ti), pi is tracing point coordinate, is determined according to tollgate devices position；ti
To be determined by the vehicle time that crosses corresponded in car data by the track point moment；N crossed car data quantity for the vehicle same day.
S2, original trip track is split as by several sub-trajectory Tr={ tr based on spectral clustering1,···,tri,···,
trm}；The purpose that sub-trajectory is split is to embody the travel behaviour each time of vehicle in original trip track.
S21, similar matrix S, the element in matrix are established using gaussian kernel function RBFWherein a, b ∈ [1, n], ρp、 ρtFor tracing point seat
Mark passes through the standard deviation of track point moment.
S22, adjacency matrix W=S, i.e., wherein element wa,b=sa,b；Degree matrix D is diagonal matrix,
S23, building Laplacian Matrix L=D-W, do standardization: D-1/2LD-1/2, most by the matrix after standardization
Feature vector corresponding to k small characteristic value forms n*k dimensional feature matrix F.
S24, the every a line composition 1*k for extracting F tie up sample matrix f, use DBSCASN clustering algorithm to whole n samples
Clustering is carried out, cluster division result { F is obtained1,···,Fi,···,Fm, wherein m is the number for the cluster that cluster generates
Amount, cluster FiIt is made of several sample matrix；According to the sample matrix in each cluster, the cluster division result of tracing point is determined
TR={ tr1,···,tri,···,trm, wherein element is single trip sub-trajectory, tri={ x(i,1),
x(i,2),···,x(i,u),···,x(i,h), wherein x(i,u)For sub-trajectory triInterior tracing point, h are the trip sub-trajectory
The tracing point quantity of approach.
S3, the space-time characteristic Variables Sequence that trip track is extracted based on discrete Fourier transform DFT；Method particularly includes:
S31, for any sub-trajectory tri={ x(i,1),x(i,2),···,x(i,u),···,x(i,h), extract track
The location dimension of point, i.e. P_tri={ p(i,1),p(i,2),···,p(i,u),···,pi,h), p(i,j)=[lng(i,u),
lat(i,u)], lng(i,u)、lat(i,u)Respectively position warp, latitude coordinate.
S32, the corresponding Fourier coefficient f_tr of the sub-trajectory is calculatedi=(fi x,fi y), whereinFor imaginary part.
S33, the Fourier coefficient F=(f_tr for generating trip track1,f_tr2,···,f_trm)。
S4, the trip track space-time characteristic Variables Sequence in historical period is obtained based on step S3, using Di Li Cray mistake
Journey mixed model carries out clustering to trip track；Di Li Cray process mixed model DPMM is a kind of topic model, this is mixed
The description of molding type are as follows:
Wherein, Dirichlet (*) is dirichlet function,For the probability of different clusters, β is that the control of clusters number is joined
Number, zlFor the l articles trip track TRlAffiliated cluster, T are trip tracking quantity to be clustered, θcFor the variance of c-th of cluster, ε is
Initial clustering number；α is the sub-trajectory number of satisfaction needed for a certain track is divided into certain cluster defined in model；xl,uFor
The l articles trip track TRlIn u-th of sub-trajectory.
Using the probability for folding Gibbs sampling method approximate solution Di Li Cray model, x and z joint probability point are obtained
Cloth function is shown below:
Wherein, P (zl=c | z-l,TR1:M, α, β) and it is the probability that the l articles trip track TRl belongs to c-th of cluster； z-lFor not
Include the track TR that goes on a journeylGathering close；TR1:MIndicate trip track bulk sample sheet；Nc(z-l) it is to reject trip track TRlC-th afterwards
It goes on a journey in cluster tracking quantity；FlFor the track TR that goes on a journeylFourier coefficient； F_columnc(TR-l) it is to be eliminated in c-th of cluster
Row track TRlThe column vector that the Fourier coefficient of other all trip tracks is constituted afterwards；Parameter in formulaWherein Γ (*) is gamma function, and τ is gamma function parameter；In order to maximize P (zl=c | z-l,
TR1:M, α, β), it is iterated sampling using Gibbs model, finally obtains the parameter Estimation of Di Li Cray process mixed model
Value.
S5, the cluster result based on step S4, using the cluster centre of each cluster as the core trip track of corresponding cluster；
The two line angle is calculated according to the coordinate of trip track starting point to the end, the traffic flow key flow direction of road network is thereby determined that, adopts
With the tracing point in K shortest path algorithm processing core trip track, K critical path is determined.
This kind of road network traffic flow key flow direction and critical path analysis method, using has higher " hold to missing information
Bear " topic model of degree analyzes track of vehicle, and then extract from track of vehicle road network traffic flow key flow direction with
Critical path provides important support for road network Demand-side signature analysis.
Embodiment method is on the basis of the sub-trajectory that spectral clustering is handled, by discrete Fourier transform to vehicle
Complete trip track is handled, its space-time characteristic variable is extracted.For clusters number can not be automatically determined in spectral clustering
Shortcoming, DBSCAN algorithm is embedded in traditional spectral clustering by embodiment method, and it is automatically true to realize optimum clustering number purpose
It is fixed.Embodiment method is based on track space-time characteristic, using Di Li Cray process mixed model, extract road network crucial flow direction and
Critical path.
Claims (6)
1. a kind of road network traffic flow key flow direction and critical path analysis method, it is characterised in that: generated based on bayonet data
Track of vehicle carries out track fractionation using Spectral Clustering, extracts the space-time characteristic variable of trip track on this basis, pass through
Topic model realizes the cluster of trip track, to identify the crucial flow direction and critical path of road network, specifically includes following steps,
S1, the history for reading tollgate devices detection cross car data, according to the brand number crossed in car data, in daily road network
All vehicles for passing through bayonet position generate the complete trip track on the vehicle same day；
S2, original trip track is split as by several sub-trajectory Tr={ tr based on spectral clustering1,…,tri,…,trm}；
S3, the space-time characteristic Variables Sequence that trip track is extracted based on discrete Fourier transform DFT；
S4, the trip track space-time characteristic Variables Sequence in historical period is obtained based on step S3, it is mixed using Di Li Cray process
Molding type carries out clustering to trip track；
S5, the cluster result based on step S4, using the cluster centre of each cluster as the core trip track of corresponding cluster；According to out
The coordinate of row track starting point to the end calculates the two line angle, thereby determines that the traffic flow key flow direction of road network；It is most short using K
Tracing point in road algorithm process core trip track, determines K critical path.
2. road network traffic flow key flow direction as described in claim 1 and critical path analysis method, it is characterised in that: step S1
In, the complete trip track on the vehicle same day is generated, specifically, what is chronologically arranged crosses vehicle point set X={ x1,…,xi,…,
xn}；Wherein xi=(pi,ti), pi is tracing point coordinate, is determined according to tollgate devices position；tiFor by the track point moment, by
The vehicle time that crosses corresponded in car data determines；N crossed car data quantity for the vehicle same day.
3. road network traffic flow key flow direction as described in claim 1 and critical path analysis method, it is characterised in that: step S1
In, it crosses car data and includes device numbering, spends vehicle time, brand number.
4. road network traffic flow key flow direction as described in any one of claims 1-3 and critical path analysis method, feature exist
In: step S2 specifically,
S21, similar matrix S, the element in matrix are established using gaussian kernel function RBFWherein a, b ∈ [1, n], ρp、ρtRespectively rail
Mark point coordinate passes through the standard deviation of track point moment；
S22, adjacency matrix W=S, i.e., wherein element wa,b=sa,b；Degree matrix D is diagonal matrix, D=diag (d1,…,di,…,
dn),
S23, building Laplacian Matrix L=D-W, do standardization: D-1/2LD-1/2, by the smallest k of matrix after standardization
Feature vector corresponding to a characteristic value forms n*k dimensional feature matrix F；
S24, the every a line composition 1*k for extracting F tie up sample matrix f, are carried out to whole n samples using DBSCASN clustering algorithm
Clustering obtains cluster division result { F1,···,Fi,···,Fm, wherein m is the quantity for the cluster that cluster generates, cluster Fi
It is made of several sample matrix；According to the sample matrix in each cluster, the cluster division result TR=of tracing point is determined
{tr1,···,tri,···,trm, wherein element is single trip sub-trajectory, tri={ x(i,1),x(i,2),···,
x(i,u),···,x(i,h), wherein x(i,u)For sub-trajectory triInterior tracing point, h are the tracing point of the trip sub-trajectory approach
Quantity.
5. road network traffic flow key flow direction as claimed in claim 4 and critical path analysis method, it is characterised in that: step S3
Specifically,
S31, for any sub-trajectory tri={ x(i,1),x(i,2),···,x(i,u),···,x(i,h), extract tracing point
Location dimension, i.e. P_tri={ p(i,1),p(i,2),···,p(i,u),···,pi,h), p(i,j)=[lng(i,u),lat(i,u)],
lng(i,u)、lat(i,u)Respectively position warp, latitude coordinate；
S32, the corresponding Fourier coefficient of the sub-trajectory is calculatedWherein
It is multiple
Number imaginary part；
S33, the Fourier coefficient F=(f_tr for generating trip track1,f_tr2,···,f_trm)。
6. road network traffic flow key flow direction as claimed in claim 5 and critical path analysis method, it is characterised in that: step S4
In, Di Li Cray process mixed model specifically:
Wherein, Dirichlet (*) is dirichlet function,For the probability of different clusters, β is the control parameter of clusters number, zl
For the l articles trip track TRlAffiliated cluster, T are trip tracking quantity to be clustered, θcFor the variance of c-th of cluster, ε is initial poly-
Class number；α is the sub-trajectory number of satisfaction needed for a certain track is divided into certain cluster defined in model；xl,uGo out for the l articles
Row track TRlIn u-th of sub-trajectory；
Using the probability for folding Gibbs sampling method approximate solution Di Li Cray model, x and z joint probability distribution function are obtained
It is shown below:
Wherein, P (zl=c | z-l,TR1:M, α, β) and it is the l articles trip track TRlBelong to the probability of c-th of cluster；z-lFor not comprising out
Row track TRlGathering close；TR1:MIndicate trip track bulk sample sheet；Nc(z-l) it is to reject trip track TRlGo out in c-th of cluster afterwards
Row tracking quantity；FlFor the track TR that goes on a journeylFourier coefficient；F_columnc(TR-l) it is that trip track is rejected in c-th of cluster
TRlThe column vector that the Fourier coefficient of other all trip tracks is constituted afterwards；Parameter in formulaWherein
Γ (*) is gamma function, and τ is gamma function parameter；In order to maximize P (zl=c | z-l,TR1:M, α, β), utilize Gibbs model
It is iterated sampling, finally obtains the estimates of parameters of Di Li Cray process mixed model.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811227838.9A CN109767615B (en) | 2018-10-19 | 2018-10-19 | Method for analyzing key flow direction and key path of road network traffic flow |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811227838.9A CN109767615B (en) | 2018-10-19 | 2018-10-19 | Method for analyzing key flow direction and key path of road network traffic flow |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109767615A true CN109767615A (en) | 2019-05-17 |
| CN109767615B CN109767615B (en) | 2021-05-18 |
Family
ID=66449868
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811227838.9A Active CN109767615B (en) | 2018-10-19 | 2018-10-19 | Method for analyzing key flow direction and key path of road network traffic flow |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109767615B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555544A (en) * | 2019-07-16 | 2019-12-10 | 广东工业大学 | traffic demand estimation method based on GPS navigation data |
| CN110827544A (en) * | 2019-11-11 | 2020-02-21 | 重庆邮电大学 | Short-term traffic flow control method based on graph convolution recurrent neural network |
| CN112102611A (en) * | 2020-08-03 | 2020-12-18 | 上海理工大学 | Urban road network key path identification method based on interruption probability influence model |
| CN113865611A (en) * | 2021-10-27 | 2021-12-31 | 北京百度网讯科技有限公司 | Data processing method, device, equipment and storage medium |
| CN114822049A (en) * | 2022-03-23 | 2022-07-29 | 山东省交通规划设计院集团有限公司 | Vehicle flow direction monitoring and analyzing method and system |
| CN115047894A (en) * | 2022-04-14 | 2022-09-13 | 中国民用航空总局第二研究所 | Unmanned aerial vehicle track measuring and calculating method, electronic equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246706A (en) * | 2013-04-09 | 2013-08-14 | 哈尔滨工程大学 | Method of clustering motion trajectories of vehicle objects in road network space |
| CN103593430A (en) * | 2013-11-11 | 2014-02-19 | 胡宝清 | Clustering method based on mobile object spatiotemporal information trajectory subsections |
| CN103605362A (en) * | 2013-09-11 | 2014-02-26 | 天津工业大学 | Learning and anomaly detection method based on multi-feature motion modes of vehicle traces |
| CN104504897A (en) * | 2014-09-28 | 2015-04-08 | 北京工业大学 | Intersection traffic flow characteristic analysis and vehicle moving prediction method based on trajectory data |
| CN104657424A (en) * | 2015-01-21 | 2015-05-27 | 段炼 | Clustering method for interest point tracks under multiple temporal and spatial characteristic fusion |
| CN105205145A (en) * | 2015-09-18 | 2015-12-30 | 中国科学院自动化研究所 | Track modeling and searching method |
| CN107204114A (en) * | 2016-03-18 | 2017-09-26 | 中兴通讯股份有限公司 | A kind of recognition methods of vehicle abnormality behavior and device |
| CN107301254A (en) * | 2017-08-24 | 2017-10-27 | 电子科技大学 | A kind of road network hot spot region method for digging |
| CN107862862A (en) * | 2016-09-22 | 2018-03-30 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle behavior analysis method and device |
-
2018
- 2018-10-19 CN CN201811227838.9A patent/CN109767615B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103246706A (en) * | 2013-04-09 | 2013-08-14 | 哈尔滨工程大学 | Method of clustering motion trajectories of vehicle objects in road network space |
| CN103605362A (en) * | 2013-09-11 | 2014-02-26 | 天津工业大学 | Learning and anomaly detection method based on multi-feature motion modes of vehicle traces |
| CN103593430A (en) * | 2013-11-11 | 2014-02-19 | 胡宝清 | Clustering method based on mobile object spatiotemporal information trajectory subsections |
| CN104504897A (en) * | 2014-09-28 | 2015-04-08 | 北京工业大学 | Intersection traffic flow characteristic analysis and vehicle moving prediction method based on trajectory data |
| CN104657424A (en) * | 2015-01-21 | 2015-05-27 | 段炼 | Clustering method for interest point tracks under multiple temporal and spatial characteristic fusion |
| CN105205145A (en) * | 2015-09-18 | 2015-12-30 | 中国科学院自动化研究所 | Track modeling and searching method |
| CN107204114A (en) * | 2016-03-18 | 2017-09-26 | 中兴通讯股份有限公司 | A kind of recognition methods of vehicle abnormality behavior and device |
| CN107862862A (en) * | 2016-09-22 | 2018-03-30 | 杭州海康威视数字技术股份有限公司 | A kind of vehicle behavior analysis method and device |
| CN107301254A (en) * | 2017-08-24 | 2017-10-27 | 电子科技大学 | A kind of road network hot spot region method for digging |
Non-Patent Citations (5)
| Title |
|---|
| HUNG C C 等: "Clustering and aggregating clues of trajectories for mining trajectory patterns and routes", 《THE VLDB JOURNAL》 * |
| 廖律超 等: "一种支持轨迹大数据潜在语义相关性挖掘的谱聚类方法", 《电子学报》 * |
| 王超 等: "基于轨迹谱聚类的终端区盛行交通流识别方法", 《西南交通大学学报》 * |
| 石陆魁 等: "基于时空模式的轨迹数据聚类算法", 《计算机应用》 * |
| 许玉萍: "基于出租车轨迹数据的城市路网车速周期性探究", 《中国优秀硕士学位论文全文数据库基础科学辑》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110555544A (en) * | 2019-07-16 | 2019-12-10 | 广东工业大学 | traffic demand estimation method based on GPS navigation data |
| CN110555544B (en) * | 2019-07-16 | 2022-04-08 | 广东工业大学 | Traffic demand estimation method based on GPS navigation data |
| CN110827544A (en) * | 2019-11-11 | 2020-02-21 | 重庆邮电大学 | Short-term traffic flow control method based on graph convolution recurrent neural network |
| CN112102611A (en) * | 2020-08-03 | 2020-12-18 | 上海理工大学 | Urban road network key path identification method based on interruption probability influence model |
| CN112102611B (en) * | 2020-08-03 | 2023-02-03 | 上海理工大学 | Urban road network key path identification method based on interruption probability influence model |
| CN113865611A (en) * | 2021-10-27 | 2021-12-31 | 北京百度网讯科技有限公司 | Data processing method, device, equipment and storage medium |
| CN114822049A (en) * | 2022-03-23 | 2022-07-29 | 山东省交通规划设计院集团有限公司 | Vehicle flow direction monitoring and analyzing method and system |
| CN114822049B (en) * | 2022-03-23 | 2023-06-20 | 山东省交通规划设计院集团有限公司 | Vehicle flow direction monitoring and analyzing method and system |
| CN115047894A (en) * | 2022-04-14 | 2022-09-13 | 中国民用航空总局第二研究所 | Unmanned aerial vehicle track measuring and calculating method, electronic equipment and storage medium |
| CN115047894B (en) * | 2022-04-14 | 2023-09-15 | 中国民用航空总局第二研究所 | Unmanned aerial vehicle track measuring and calculating method, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109767615B (en) | 2021-05-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109767615A (en) | Road network traffic flow key flow direction and critical path analysis method | |
| Karagiorgou et al. | On vehicle tracking data-based road network generation | |
| CN109448370B (en) | Traffic control subarea division method based on vehicle track data | |
| CN104330089B (en) | A kind of method that map match is carried out using history gps data | |
| CN109871876A (en) | A kind of Freeway Conditions identification and prediction technique based on floating car data | |
| CN109410586A (en) | A kind of Traffic State Detection Method based on multivariate data fusion | |
| CN110288032B (en) | Vehicle driving track type detection method and device | |
| CN108959466A (en) | Taxi hot spot method for visualizing and system based on BCS-DBSCAN | |
| CN106650913A (en) | Deep convolution neural network-based traffic flow density estimation method | |
| CN105513370B (en) | The traffic zone division methods excavated based on sparse license plate identification data | |
| CN106710208A (en) | Traffic state acquisition method and device | |
| CN106446208A (en) | Method for identifying traveling mode of smart phone by considering integrating degree of road network | |
| CN106097717B (en) | Signalized intersections based on the fusion of two class floating car datas are averaged transit time method of estimation | |
| Zhu et al. | Inferring taxi status using gps trajectories | |
| CN109254861A (en) | OD requirement extract and its analysis method for reliability based on track data | |
| CN109359690A (en) | Vehicle driving track recognizing method based on bayonet data | |
| CN108171976A (en) | Based on the vehicular applications recognition methods in way data | |
| CN110443288A (en) | A kind of track similarity calculation method based on sequence study | |
| Lin et al. | Vehicle re-identification with dynamic time windows for vehicle passage time estimation | |
| CN102968900A (en) | Method for processing RFID (Radio Frequency Identification) traffic data | |
| CN106203296B (en) | The video actions recognition methods of one attribute auxiliary | |
| CN110516633A (en) | A kind of method for detecting lane lines and system based on deep learning | |
| CN111949896A (en) | Track anomaly detection method based on track big data | |
| CN107274679A (en) | Vehicle identification method, device, equipment and computer-readable recording medium | |
| CN109816271A (en) | Cycle track service level evaluation method based on shared bicycle track data |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 211106 19 Su Yuan Avenue, Jiangning economic and Technological Development Zone, Nanjing, Jiangsu Applicant after: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. Address before: 210006, Qinhuai District, Jiangsu, Nanjing should be 388 days street, Chenguang 1865 Technology Creative Industry Park E10 building on the third floor Applicant before: JIANGSU ZHITONG TRAFFIC TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |