CN108492557A - Highway jam level judgment method based on multi-model fusion - Google Patents
Highway jam level judgment method based on multi-model fusion Download PDFInfo
- Publication number
- CN108492557A CN108492557A CN201810255202.9A CN201810255202A CN108492557A CN 108492557 A CN108492557 A CN 108492557A CN 201810255202 A CN201810255202 A CN 201810255202A CN 108492557 A CN108492557 A CN 108492557A
- Authority
- CN
- China
- Prior art keywords
- model
- sample
- data
- jam level
- highway
- 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.)
- Pending
Links
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
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- 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
- G08G1/0133—Traffic data processing for classifying traffic situation
Abstract
The invention discloses a kind of highway jam level judgment methods based on multi-model fusion comprising following steps：S1, highway flow histories data are obtained and are normalized；S2, jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data, obtain the jam level of historical data, i.e. judgment models；S3, the jam level according to historical data carry out classification analysis to the highway flow real time data after normalized simultaneously using different sorting algorithms, obtain multiple jam levels of real time data, i.e. disaggregated model；S4, disaggregated model is merged by judgment models, obtains the real-time jam level of highway.The present invention, as model, is simultaneously trained model using a variety of sorting algorithms, merges that each model obtains as a result, effectively improving discrimination precision using historical data.
Description
Technical field
The present invention relates to data analysis fields, and in particular to it is a kind of based on multi-model fusion highway jam level sentence
Disconnected method.
Background technology
With the quickening of China's rapid development of economy and social progress step, the traffic capacity of freeway network is
Growing transport need is cannot be satisfied, congested in traffic and clogging becomes increasingly severe, and traffic jam level differentiates
The service level that road network can be reacted in real time from global angle is the weight that traffic control system is cooperateed with traffic flow guidance system
It will foundation.Traffic jam level differentiates while reflection traffic flow objective operating status, can be provided to traffic administration person accurate
True ground traffic noise prediction information.When in this differentiation result being provided to transportation system management person and policymaker's hand, they are just
Different situations can be directed to and formulate corresponding traffic control, management and induction measure.In addition, the differentiation of traffic behavior can be to road
The collected all kinds of traffic datas of roadbed plinth information acquisition system are analyzed, and current traffic is obtained according to their variation tendency
The operation conditions of system.
Traffic flow average travel speed data are mainly obtained for the differentiation of jam level at present, by by itself and threshold value
It is compared, to realize the differentiation to traffic status of express way.But this method be only applicable to distinguish it is crowded or uncongested two
Kind traffic behavior, if traffic behavior to be carried out to more careful division, the accuracy of this method will decline very much.In addition,
At present in terms of the differentiation of traffic behavior, it is concentrated mainly on the judgement to urban road congestion grade, for the friendship of highway
, not yet there is the system discrimination standard of a set of maturation in logical condition discrimination, this, which also gives, carries out differentiation using the above method and bring at present
No small difficulty.
Invention content
For above-mentioned deficiency in the prior art, a kind of highway congestion based on multi-model fusion provided by the invention
Level determination method solves the problems, such as that existing jam level discriminant accuracy is low.
In order to reach foregoing invention purpose, the technical solution adopted by the present invention is：
There is provided a kind of highway jam level judgment method merged based on multi-model comprising following steps：
S1, highway flow histories data are obtained and are normalized；
S2, jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data,
Obtain the jam level of historical data, i.e. judgment models；
S3, the jam level according to historical data, using different sorting algorithms simultaneously to the high speed after normalized
Highway flow real time data carries out classification analysis, obtains multiple jam levels of real time data, i.e. disaggregated model；
S4, disaggregated model is merged by judgment models, obtains the real-time jam level of highway.
Further, highway flow histories data are obtained and the method being normalized is：
The historical data for obtaining freeway traffic flow road network operating status, is arranged corresponding time window, by time window
Historical data in mouthful collects the single unit vehicle travel direction traffic flow data for obtaining time interval T, it is normalized place
Reason；Wherein traffic flow data includes：The magnitude of traffic flow, travel speed and occupation rate.
Further, jam level number K is preset according to the data after normalized, and it is poly- to carry out K mean values to data
Alanysis, obtains the jam level of historical data, i.e. the method for judgment models is：
S2-1, according to the data acquisition training sample after normalized
{x(1),x(2),…x(m)},xi∈Rn
Wherein xiFor individualized training sample, i=1,2 ..., m；RnFor total training sample；
S2-2, setting jam level number K, are randomly derived K cluster centre point：
μ1,μ2,…μk∈Rn
S2-3, according to formula
The classification of each sample i is obtained, and according to formula
Update is per a kind of barycenter μj, until distortion function
Convergence obtains the K barycenter no longer changed；
S2-4, using the minimum value of sample and each cluster centre point distance as the sample belong to the judgement of certain classification according to
According to i.e. judgment models；
Wherein K is given cluster numbers, C(i)Representative sample i and the nearest classification of distance, C in k class(i)=1,2 ..., K；
Barycenter μjThe conjecture to the center of a sample's point for belonging to same class is represented, after distortion function convergence, μjAs in such particle
The heart；J=0,1,2 ..., k；J (c, μ) is distortion function, μcFor the center after the completion of cluster.
Further, K values are 4, that is, are divided into the grade of 4 congestion levels from low to high：Unimpeded, jogging, congestion and tight
Congestion again.
Further, disaggregated model includes model-naive Bayesian, decision-tree model, SVM models and Logic Regression Models：
The sorting technique of model-naive Bayesian is：
S3-1-1, according to the data acquisition sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-1-2, the sample according to acquisition and following formula：
Obtain some value a of sample characteristicsjlGiving certain classification c(k)Under Probability p (xj=ajl| y=c(k)), i.e. mould
The elementary probability of type；Wherein I (x) is indicator function, is calculated as 1 if being set up in bracket, is otherwise 0；N indicates total sample number, i=
1,2,…,N；
S3-1-3, the elementary probability according to model, pass through formula
Obtain giving the probability of unfiled new instance X；
S3-1-4, according to formula
Obtain the classification y described in the instance X；
The sorting technique of decision-tree model is：
S3-2-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-2-2, according to formula
Entropy H is obtained, wherein x is classification, and p (x) is the probability that any sample is the classification；
S3-2-3, data set is divided according to the maximum principle of entropy difference；
The sorting technique of SVM models is：
S3-3-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-3-2, according to formula
k(x,xi)=((x*xi)+1)d
Classified and obtains disaggregated model, wherein xiIndicate that i-th of sample, d are constant；
The sorting technique of Logic Regression Models is：
S3-4-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-4-2, according to sigmoid functions
Z=θ0+θ1x1+θ2x2+θ3x3+…+θnxn
Classified and obtains disaggregated model y；Wherein e is natural constant；θ0, θ1, θ2…θnIt is constant；x1, x2,
x3... xnRespectively first sample, second sample, n-th of sample of third sample ....
Further, disaggregated model is merged by judgment models, the method for obtaining highway jam level is：
S4-1, it will determine that the data of model are randomly divided into training set and test set according to a certain percentage；
S4-2, disaggregated model is trained according to training set, obtains the model after each fusion；
S4-3, test set is tested according to the model after each fusion, obtains the precision of model after each fusion；
S4-4, it at least repeats that step S4-1 is primary to step S4-3, obtains the mean accuracy G of model after each fusionb, b=
1,2 ..., n, n are the sum of model after fusion；
S4-5, according to formula
Obtain the weight P of model mean accuracy after each fusionn；
S4-6, according to formula
G=P1G1+P2G2+P3G3+…++PnGn
The final result G of model after each fusion is obtained, according to the minimum of final result G and each jam level barycenter numerical value
Difference judges jam level；
Wherein M is model training number, NCThe quantity summation hit using the disaggregated model for M times, NAFor test set sample
Sum.
Further, the ratio of training set and test set is 3:1.
Beneficial effects of the present invention are：The present invention based on the data that freeway traffic collecting device obtains, with when
Between in window by vehicle flowrate quantity, section average speed and occupation rate etc. historical data is gathered for core parameter
Class, and using result as the foundation of real time data progress classification learning.Finally merge it is that a variety of classification learning models obtain as a result,
Real time discriminating is carried out to jam level, effectively increases the discrimination precision to jam level.
Description of the drawings
Fig. 1 is the flow chart of the present invention.
Specific implementation mode
The specific implementation mode of the present invention is described below, in order to facilitate understanding by those skilled in the art this hair
It is bright, it should be apparent that the present invention is not limited to the ranges of specific implementation mode, for those skilled in the art,
As long as various change is in the spirit and scope of the present invention that the attached claims limit and determine, these variations are aobvious and easy
See, all are using the innovation and creation of present inventive concept in the row of protection.
As shown in Figure 1, should be included the following steps based on the highway jam level judgment method that multi-model merges：
S1, highway flow histories data are obtained and are normalized；
S2, jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data,
Obtain the jam level of historical data, i.e. judgment models；
S3, the jam level according to historical data, using different sorting algorithms simultaneously to the high speed after normalized
Highway flow real time data carries out classification analysis, obtains multiple jam levels of real time data, i.e. disaggregated model；
S4, disaggregated model is merged by judgment models, obtains the real-time jam level of highway.
It obtains highway flow histories data and the method being normalized is：
The historical data for obtaining freeway traffic flow road network operating status, is arranged corresponding time window, by time window
Historical data in mouthful collects the single unit vehicle travel direction traffic flow data for obtaining time interval T, it is normalized place
Reason；Wherein traffic flow data includes：The magnitude of traffic flow, space headway, travel speed, traffic density, queue length and occupation rate.
Jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data, is obtained
Method to the jam level of historical data, i.e. judgment models is：
S2-1, according to the data acquisition training sample after normalized
{x(1),x(2),…x(m)},xi∈Rn
Wherein xiFor individualized training sample, i=1,2 ..., m；RnFor total training sample；
S2-2, setting jam level number K, are randomly derived K cluster centre point：
μ1,μ2,…μk∈Rn
S2-3, according to formula
The classification of each sample i is obtained, and according to formula
Update is per a kind of barycenter μj, until distortion function
Convergence obtains the K barycenter no longer changed；
S2-4, using the minimum value of sample and each cluster centre point distance as the sample belong to the judgement of certain classification according to
According to i.e. judgment models；
Wherein K is given cluster numbers, C(i)Representative sample i and the nearest classification of distance, C in k class(i)=1,2 ..., K；
Barycenter μjThe conjecture to the center of a sample's point for belonging to same class is represented, after distortion function convergence, μjAs in such particle
The heart；J=0,1,2 ..., k；J (c, μ) is distortion function, μcFor the center after the completion of cluster.
K values are 4, that is, are divided into the grade of 4 congestion levels from low to high：Unimpeded, jogging, congestion and heavy congestion.
Disaggregated model includes model-naive Bayesian, decision-tree model, SVM models and Logic Regression Models：
The sorting technique of model-naive Bayesian is：
S3-1-1, according to the data acquisition sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-1-2, the sample according to acquisition and following formula：
Obtain some value a of sample characteristicsjlGiving certain classification c(k)Under Probability p (xj=ajl| y=c(k)), i.e. mould
The elementary probability of type；Wherein I (x) is indicator function, is calculated as 1 if being set up in bracket, is otherwise 0；N indicates total sample number, i=
1,2,…,N；
S3-1-3, the elementary probability according to model, pass through formula
Obtain giving the probability of unfiled new instance X；
S3-1-4, according to formula
Obtain the classification y described in the instance X；
The sorting technique of decision-tree model is：
S3-2-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-2-2, according to formula
Entropy H is obtained, wherein x is classification, and p (x) is the probability that any sample is the classification；
S3-2-3, data set is divided according to the maximum principle of entropy difference；
The sorting technique of the SVM models is：
S3-3-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-3-2, according to formula
k(x,xi)=((x*xi)+1)d
Classified and obtains disaggregated model, wherein xiIndicate that i-th of sample, d are constant；
The sorting technique of Logic Regression Models is：
S3-4-1, according to the data acquisition training sample after normalized：
{(x1,y1),(x2,y2),…(xN,yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；
yi∈(c(1),c(2),…c(k))TBelong to one kind in K classes；
S3-4-2, according to sigmoid functions
Z=θ0+θ1x1+θ2x2+θ3x3+…+θnxn
Classified and obtains disaggregated model y；Wherein e is natural constant；θ0, θ1, θ2…θnIt is constant；x1, x2,
x3... xnRespectively first sample, second sample, n-th of sample of third sample ....
Disaggregated model is merged by judgment models, the method for obtaining highway jam level is：
S4-1, it will determine that the data of model are randomly divided into training set and test set according to a certain percentage；
S4-2, disaggregated model is trained according to training set, obtains the model after each fusion；
S4-3, test set is tested according to the model after each fusion, obtains the precision of model after each fusion；
S4-4, it at least repeats that step S4-1 is primary to step S4-3, obtains the mean accuracy G of model after each fusionb, b=
1,2 ..., n, n are the sum of model after fusion；
S4-5, according to formula
Obtain the weight P of model mean accuracy after each fusionn；
S4-6, according to formula
G=P1G1+P2G2+P3G3+…++PnGn
The final result G of model after each fusion is obtained, according to the minimum of final result G and each jam level barycenter numerical value
Difference judges jam level；
Wherein M is model training number, NCThe quantity summation hit using the disaggregated model for M times, NAFor test set sample
Sum.
The ratio of training set and test set is 3:1.
In conclusion the present invention is based on the data that freeway traffic collecting device obtains, to lead in time window
The average speed of the vehicle flowrate quantity, section crossed and occupation rate etc. are that core parameter clusters historical data, and will tie
Fruit carries out the foundation of classification learning as real time data.It finally merges that a variety of classification learning models obtain as a result, to congestion etc.
Grade carries out real time discriminating, effectively increases the discrimination precision to jam level.
Claims (7)
1. a kind of highway jam level judgment method based on multi-model fusion, which is characterized in that include the following steps：
S1, highway flow histories data are obtained and are normalized；
S2, jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data, obtain
The jam level of historical data, i.e. judgment models；
S3, the jam level according to historical data, using different sorting algorithms simultaneously to the highway after normalized
Flow real time data carries out classification analysis, obtains multiple jam levels of real time data, i.e. disaggregated model；
S4, disaggregated model is merged by judgment models, obtains the real-time jam level of highway.
2. the highway jam level judgment method according to claim 1 based on multi-model fusion, which is characterized in that
It obtains highway flow histories data and the method being normalized is：
The historical data for obtaining freeway traffic flow road network operating status, is arranged corresponding time window, will be in time window
Historical data collect the single unit vehicle travel direction traffic flow data for obtaining time interval T, it is normalized；Its
Middle traffic flow data includes：The magnitude of traffic flow, travel speed and occupation rate.
3. the highway jam level judgment method according to claim 2 based on multi-model fusion, which is characterized in that
Jam level number K is preset according to the data after normalized, and K mean cluster analysis is carried out to data, obtains history number
According to jam level, i.e. the method for judgment models is：
S2-1, according to the data acquisition training sample after normalized
{x(1), x(2)... x(m), xi∈Rn
Wherein xiFor individualized training sample, i=1,2 ..., m；RnFor total training sample；
S2-2, setting jam level number K, are randomly derived K cluster centre point：
μ1, μ2... μk∈Rn
S2-3, according to formula
The classification of each sample i is obtained, and according to formula
Update is per a kind of barycenter μj, until distortion function
Convergence obtains the K barycenter no longer changed；
S2-4, the basis for estimation for belonging to certain classification using the minimum value of sample and each cluster centre point distance as the sample, i.e.,
Judgment models；
Wherein K is given cluster numbers, C(i)Representative sample i and the nearest classification of distance, C in k class(i)=1,2 ..., K；Barycenter μj
The conjecture to the center of a sample's point for belonging to same class is represented, after distortion function convergence, μjAs such particle center；J=
0,1,2 ..., k；J (c, μ) is distortion function, μcFor the center after the completion of cluster.
4. the highway jam level judgment method according to claim 3 based on multi-model fusion, it is characterised in that：
The K values are 4, that is, are divided into the grade of 4 congestion levels from low to high：Unimpeded, jogging, congestion and heavy congestion.
5. the highway jam level judgment method according to claim 3 based on multi-model fusion, which is characterized in that
The disaggregated model includes model-naive Bayesian, decision-tree model, SVM models and Logic Regression Models：
The sorting technique of the model-naive Bayesian is：
S3-1-1, according to the data acquisition sample after normalized：
{(x1, y1), (x2, y2) ... (xN, yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；yi∈
(c(1), c(2)... c(k))TBelong to one kind in K classes；
S3-1-2, the sample according to acquisition and following formula：
Obtain some value a of sample characteristicsj1Giving certain classification c(k)Under Probability p (xj=ajl| y=c(k)), i.e. model
Elementary probability；Wherein I (x) is indicator function, is calculated as 1 if being set up in bracket, is otherwise 0；N expression total sample numbers, i=1,
2 ..., N；
S3-1-3, the elementary probability according to model, pass through formula
Obtain giving the probability of unfiled new instance X；
S3-1-4, according to formula
Obtain the classification y described in the instance X；
The sorting technique of the decision-tree model is：
S3-2-1, according to the data acquisition training sample after normalized：
{(x1, y1), (x2, y2) ... (xN, yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；yi∈
(c(1), c(2)... c(k))TBelong to one kind in K classes；
S3-2-2, according to formula
Entropy H is obtained, wherein x is classification, and p (x) is the probability that any sample is the classification；
S3-2-3, data set is divided according to the maximum principle of entropy difference；
The sorting technique of the SVM models is：
S3-3-1, according to the data acquisition training sample after normalized：
{(x1, y1), (x2, y2) ... (xN, yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；yi∈
(c(1), c(2)... c(k))TBelong to one kind in K classes；
S3-3-2, according to formula
K (x, xi)=((x*xi)+1)d
Classified and obtains disaggregated model, wherein xiIndicate that i-th of sample, d are constant；
The sorting technique of the Logic Regression Models is：
S3-4-1, according to the data acquisition training sample after normalized：
{(x1, y1), (x2, y2) ... (xN, yN)}
WhereinIt is M dimensional vectors, including the magnitude of traffic flow, travel speed and occupation rate；yi∈
(c(1), c(2)... c(k))TBelong to one kind in K classes；
S3-4-2, according to sigmoid functions
Z=θ0+θ1x1+θ2x2+θ3x3+…+θnxn
Classified and obtains disaggregated model y；Wherein e is natural constant；θ0, θ1, θ2…θnIt is constant；x1, x2, x3... xnPoint
Not Wei first sample, second sample, n-th of sample of third sample ....
6. the highway jam level judgment method according to claim 5 based on multi-model fusion, which is characterized in that
Disaggregated model is merged by judgment models, the method for obtaining highway jam level is：
S4-1, it will determine that the data of model are randomly divided into training set and test set according to a certain percentage；
S4-2, disaggregated model is trained according to training set, obtains the model after each fusion；
S4-3, test set is tested according to the model after each fusion, obtains the precision of model after each fusion；
S4-4, it at least repeats that step S4-1 is primary to step S4-3, obtains the mean accuracy G of model after each fusionb, b=1,
2 ..., n, n are the sum of model after fusion；
S4-5, according to formula
Obtain the weight P of model mean accuracy after each fusionn；
S4-6, according to formula
G=P1G1+P2G2+P3G3+…++PnGn
The final result G of model after each fusion is obtained, according to the minimal difference of final result G and each jam level barycenter numerical value
Judge jam level；
Wherein M is model training number, NCThe quantity summation hit using the disaggregated model for M times, NAIt is total for test set sample
Number.
7. the highway jam level judgment method according to claim 6 based on multi-model fusion, it is characterised in that：
The ratio of training set and test set is 3: 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810255202.9A CN108492557A (en) | 2018-03-23 | 2018-03-23 | Highway jam level judgment method based on multi-model fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810255202.9A CN108492557A (en) | 2018-03-23 | 2018-03-23 | Highway jam level judgment method based on multi-model fusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108492557A true CN108492557A (en) | 2018-09-04 |
Family
ID=63338013
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810255202.9A Pending CN108492557A (en) | 2018-03-23 | 2018-03-23 | Highway jam level judgment method based on multi-model fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108492557A (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109242209A (en) * | 2018-10-12 | 2019-01-18 | 北京交通大学 | Railway emergency event grading forewarning system method based on K-means cluster |
| CN109389178A (en) * | 2018-10-26 | 2019-02-26 | 深圳市元征科技股份有限公司 | A kind of maintenance factory's ranking method, system and electronic equipment and storage medium |
| CN109446881A (en) * | 2018-09-05 | 2019-03-08 | 重庆大学 | A kind of express highway section Traffic State Detection Method based on isomeric data |
| CN109859477A (en) * | 2019-03-15 | 2019-06-07 | 同盾控股有限公司 | A kind of determination method and apparatus of congestion data |
| CN109871412A (en) * | 2018-12-26 | 2019-06-11 | 航天科工广信智能技术有限公司 | Lane flow analysis method based on K-Means cluster |
| CN110210509A (en) * | 2019-03-04 | 2019-09-06 | 广东交通职业技术学院 | A kind of road net traffic state method of discrimination based on MFD+ spectral clustering+SVM |
| US10453337B2 (en) * | 2015-06-25 | 2019-10-22 | Here Global B.V. | Method and apparatus for providing safety levels estimate for a travel link based on signage information |
| CN110390816A (en) * | 2019-07-18 | 2019-10-29 | 电子科技大学 | A kind of condition discrimination method based on multi-model fusion |
| CN110399364A (en) * | 2019-07-11 | 2019-11-01 | 北京航空航天大学 | A kind of data fusion method based on a variety of highway detector datas |
| CN110516702A (en) * | 2019-07-18 | 2019-11-29 | 电子科技大学 | A kind of discreet paths planing method based on flow data |
| CN110796867A (en) * | 2019-11-28 | 2020-02-14 | 沈阳世纪高通科技有限公司 | Road condition determining method and device |
| CN111209934A (en) * | 2019-12-26 | 2020-05-29 | 大唐新疆清洁能源有限公司 | Fan fault prediction and alarm method and system |
| CN113034892A (en) * | 2019-12-25 | 2021-06-25 | 南京理工大学 | Highway congestion identification method based on dynamic data stream real-time trend analysis |
| WO2021169174A1 (en) * | 2020-02-29 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Road congestion degree prediction method, apparatus, computer device, and readable storage medium |
| CN114005275A (en) * | 2021-10-25 | 2022-02-01 | 浙江交投高速公路运营管理有限公司 | Highway vehicle congestion judging method based on multi-data source fusion |
| WO2023206166A1 (en) * | 2022-04-27 | 2023-11-02 | 华为技术有限公司 | Object detection method and apparatus |
| CN117152973A (en) * | 2023-10-27 | 2023-12-01 | 贵州宏信达高新科技有限责任公司 | Expressway real-time flow monitoring method and system based on ETC portal data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101599217A (en) * | 2009-07-17 | 2009-12-09 | 北京交通大学 | A kind of method for rapidly judging traffic state |
| CN102750824A (en) * | 2012-06-19 | 2012-10-24 | 银江股份有限公司 | Urban road traffic condition detection method based on voting of network sorter |
| CN103150900A (en) * | 2013-02-04 | 2013-06-12 | 合肥革绿信息科技有限公司 | Traffic jam event automatic detecting method based on videos |
| CN103839409A (en) * | 2014-02-27 | 2014-06-04 | 南京大学 | Traffic flow state judgment method based on multiple-cross-section vision sensing clustering analysis |
| CN106211084A (en) * | 2016-09-07 | 2016-12-07 | 中国人民解放军国防科学技术大学 | Environment perception method based on GSM signal |
| CN106652460A (en) * | 2017-03-08 | 2017-05-10 | 吉林大学 | Highway traffic state determination method and system |
-
2018
- 2018-03-23 CN CN201810255202.9A patent/CN108492557A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101599217A (en) * | 2009-07-17 | 2009-12-09 | 北京交通大学 | A kind of method for rapidly judging traffic state |
| CN102750824A (en) * | 2012-06-19 | 2012-10-24 | 银江股份有限公司 | Urban road traffic condition detection method based on voting of network sorter |
| CN103150900A (en) * | 2013-02-04 | 2013-06-12 | 合肥革绿信息科技有限公司 | Traffic jam event automatic detecting method based on videos |
| CN103839409A (en) * | 2014-02-27 | 2014-06-04 | 南京大学 | Traffic flow state judgment method based on multiple-cross-section vision sensing clustering analysis |
| CN106211084A (en) * | 2016-09-07 | 2016-12-07 | 中国人民解放军国防科学技术大学 | Environment perception method based on GSM signal |
| CN106652460A (en) * | 2017-03-08 | 2017-05-10 | 吉林大学 | Highway traffic state determination method and system |
Non-Patent Citations (3)
| Title |
|---|
| 焦李成 等: "《自然计算、机器学习与图像理解前沿》", 31 August 2008, 西安电子科技大学出版社 * |
| 陈新泉: "《聚类算法中的优化方法应用》", 31 July 2014, 电子科技大学出版社 * |
| 韩露莎: "ITS中交通状态检测的算法研究", 《中国优秀硕士学位论文全文数据库·工程科技Ⅱ辑》 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10453337B2 (en) * | 2015-06-25 | 2019-10-22 | Here Global B.V. | Method and apparatus for providing safety levels estimate for a travel link based on signage information |
| CN109446881B (en) * | 2018-09-05 | 2022-06-24 | 重庆大学 | Heterogeneous data-based highway section traffic state detection method |
| CN109446881A (en) * | 2018-09-05 | 2019-03-08 | 重庆大学 | A kind of express highway section Traffic State Detection Method based on isomeric data |
| CN109242209A (en) * | 2018-10-12 | 2019-01-18 | 北京交通大学 | Railway emergency event grading forewarning system method based on K-means cluster |
| CN109389178A (en) * | 2018-10-26 | 2019-02-26 | 深圳市元征科技股份有限公司 | A kind of maintenance factory's ranking method, system and electronic equipment and storage medium |
| CN109871412A (en) * | 2018-12-26 | 2019-06-11 | 航天科工广信智能技术有限公司 | Lane flow analysis method based on K-Means cluster |
| CN110210509A (en) * | 2019-03-04 | 2019-09-06 | 广东交通职业技术学院 | A kind of road net traffic state method of discrimination based on MFD+ spectral clustering+SVM |
| CN109859477A (en) * | 2019-03-15 | 2019-06-07 | 同盾控股有限公司 | A kind of determination method and apparatus of congestion data |
| CN110399364A (en) * | 2019-07-11 | 2019-11-01 | 北京航空航天大学 | A kind of data fusion method based on a variety of highway detector datas |
| CN110390816A (en) * | 2019-07-18 | 2019-10-29 | 电子科技大学 | A kind of condition discrimination method based on multi-model fusion |
| CN110516702A (en) * | 2019-07-18 | 2019-11-29 | 电子科技大学 | A kind of discreet paths planing method based on flow data |
| CN110796867A (en) * | 2019-11-28 | 2020-02-14 | 沈阳世纪高通科技有限公司 | Road condition determining method and device |
| CN110796867B (en) * | 2019-11-28 | 2021-04-16 | 沈阳世纪高通科技有限公司 | Road condition determining method and device |
| CN113034892A (en) * | 2019-12-25 | 2021-06-25 | 南京理工大学 | Highway congestion identification method based on dynamic data stream real-time trend analysis |
| CN113034892B (en) * | 2019-12-25 | 2022-09-06 | 南京理工大学 | Highway congestion identification method based on dynamic data stream real-time trend analysis |
| CN111209934A (en) * | 2019-12-26 | 2020-05-29 | 大唐新疆清洁能源有限公司 | Fan fault prediction and alarm method and system |
| CN111209934B (en) * | 2019-12-26 | 2023-11-21 | 大唐新疆清洁能源有限公司 | Fan fault pre-alarm method and system |
| WO2021169174A1 (en) * | 2020-02-29 | 2021-09-02 | 深圳壹账通智能科技有限公司 | Road congestion degree prediction method, apparatus, computer device, and readable storage medium |
| CN114005275A (en) * | 2021-10-25 | 2022-02-01 | 浙江交投高速公路运营管理有限公司 | Highway vehicle congestion judging method based on multi-data source fusion |
| CN114005275B (en) * | 2021-10-25 | 2023-03-14 | 浙江交投高速公路运营管理有限公司 | Highway vehicle congestion judging method based on multi-data source fusion |
| WO2023206166A1 (en) * | 2022-04-27 | 2023-11-02 | 华为技术有限公司 | Object detection method and apparatus |
| CN117152973A (en) * | 2023-10-27 | 2023-12-01 | 贵州宏信达高新科技有限责任公司 | Expressway real-time flow monitoring method and system based on ETC portal data |
| CN117152973B (en) * | 2023-10-27 | 2024-01-05 | 贵州宏信达高新科技有限责任公司 | Expressway real-time flow monitoring method and system based on ETC portal data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108492557A (en) | Highway jam level judgment method based on multi-model fusion | |
| CN109544932B (en) | Urban road network flow estimation method based on fusion of taxi GPS data and gate data | |
| CN108492555A (en) | A kind of city road net traffic state evaluation method and device | |
| CN104809877B (en) | The highway place traffic state estimation method of feature based parameter weighting GEFCM algorithms | |
| CN104778837B (en) | A kind of road traffic operation situation Multiple Time Scales Forecasting Methodology | |
| CN105913661B (en) | A kind of express highway section traffic state judging method based on charge data | |
| CN100444210C (en) | Mixed controlling method of single dot signal controlling crossing | |
| CN104021671B (en) | The determination methods of the road real-time road that a kind of svm combines with fuzzy Judgment | |
| CN106530715B (en) | Road net traffic state prediction technique based on fuzzy Markov process | |
| CN108364467A (en) | A kind of traffic information prediction technique based on modified decision Tree algorithms | |
| CN108091132B (en) | Traffic flow prediction method and device | |
| CN107492251A (en) | It is a kind of to be identified and driving condition supervision method based on the driver identity of machine learning and deep learning | |
| CN101188064B (en) | 3D integrated freeway traffic event automatic detection method | |
| CN105355049B (en) | A kind of highway evaluation of running status method based on macroscopical parent map | |
| CN109686090B (en) | Virtual traffic flow calculation method based on multi-source data fusion | |
| CN107507421B (en) | Method for rapidly judging traffic state and device | |
| CN110085026A (en) | A kind of traffic status prediction method based on clustering and Markov model | |
| CN113436433B (en) | Efficient urban traffic outlier detection method | |
| CN109272745A (en) | A kind of track of vehicle prediction technique based on deep neural network | |
| CN109489679B (en) | Arrival time calculation method in navigation path | |
| CN109118787A (en) | A kind of car speed prediction technique based on deep neural network | |
| CN105551250A (en) | Method for discriminating urban road intersection operation state on the basis of interval clustering | |
| Oh et al. | Real-time inductive-signature-based level of service for signalized intersections | |
| CN109712398A (en) | Motorway journeys time Estimate Model Parameter Optimization method | |
| CN111063204A (en) | Expressway vehicle speed prediction model training method based on toll station flow |
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 | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Ken Inventor after: Chen Xueqin Inventor after: Gao Qiang Inventor after: Liu Juexiong Inventor after: Di Jiayi Inventor after: Tang Chen Inventor after: Wang Bintao Inventor after: Li Kai Inventor after: Chen Fei Inventor after: Fan Yong Inventor after: Yang Wanyi Inventor after: Wang Ruijin Inventor after: Zhang Fengli Inventor after: Zhang Xueyan Inventor after: Jiang Guichuan Inventor before: Chen Fei Inventor before: Liu Juexiong Inventor before: Di Jiayi Inventor before: Tang Chen Inventor before: Wang Bintao Inventor before: Wang Ruijin Inventor before: Li Kai Inventor before: Zhang Fengli Inventor before: Yang Wanyi Inventor before: Zhang Xueyan Inventor before: Jiang Guichuan Inventor before: Chen Xueqin Inventor before: Gao Qiang |
|
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Chen Fei Inventor after: Liu Juexiong Inventor after: Zhai Jiayi Inventor after: Tang Chen Inventor after: Wang Bintao Inventor after: Wang Ruijin Inventor after: Li Kai Inventor after: Zhang Fengli Inventor after: Yang Wanyi Inventor after: Zhang Xueyan Inventor after: Jiang Guichuan Inventor after: Chen Xueqin Inventor after: Gao Qiang Inventor before: Chen Ken Inventor before: Chen Xueqin Inventor before: Gao Qiang Inventor before: Liu Juexiong Inventor before: Zhai Jiayi Inventor before: Tang Chen Inventor before: Wang Bintao Inventor before: Li Kai Inventor before: Chen Fei Inventor before: Fan Yong Inventor before: Yang Wanyi Inventor before: Wang Ruijin Inventor before: Zhang Fengli Inventor before: Zhang Xueyan Inventor before: Jiang Guichuan |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180904 |