CN103488878A - Vector similarity based traffic flow time sequence change point identification method - Google Patents
Vector similarity based traffic flow time sequence change point identification method Download PDFInfo
- Publication number
- CN103488878A CN103488878A CN201310406641.2A CN201310406641A CN103488878A CN 103488878 A CN103488878 A CN 103488878A CN 201310406641 A CN201310406641 A CN 201310406641A CN 103488878 A CN103488878 A CN 103488878A
- Authority
- CN
- China
- Prior art keywords
- traffic flow
- similarity
- height
- time series
- tfv
- 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
Images
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention relates to the field of traffic flow state analysis, in particular to a vector similarity based traffic flow time sequence change point identification method. Through the method, sudden change of traffic flow states and change point positions of traffic flow can be judged to prevent possible traffic abnormalities in advance. The method includes the following steps: acquiring section traffic flow parameters of a detected road segment through a detector, and storing the section traffic flow parameters to a traffic flow parameter database; reading the traffic flow parameters from the traffic flow parameter database according to a time sequence, and creating a traffic flow vector set of a traffic flow time sequence; according to the traffic flow vector set, establishing a vector similarity function of the traffic flow time sequence so as to obtain similarity of traffic flow parameter evolvement rules before and after each time point; identifying type of each point in the traffic flow time sequence, then judging whether the traffic flow states are changed suddenly or not, and identifying the occurrence position of the sudden change.
Description
Technical field
The present invention relates to the traffic flow modes analysis field, relate in particular to the recognition methods of a kind of Traffic Flow Time Series height.
Background technology
Along with the sharp increase of automobile pollution, road load increases the weight of, and traffic is supplied with and can not be met the transport need of day by day expanding, and congested in traffic, frequent accidents occurs.This not only incurs loss through delay the journey time of traveler, reduces traffic circulation efficiency, upsets normal traffic order, and contaminated environment, waste energy, and when serious, also may threaten the life security of traveler.Therefore, analyze traffic flow modes, in time, identification traffic conditions exactly, grasp the traffic flow Evolution Characteristics, for formulating rationally effectively traffic guidance and controlling significant.
Research for the traffic flow modes analysis system in the past mainly concentrates in the real-time identification and pre-identification of traffic behavior, focuses on the qualitative change of traffic behavior, and has ignored the research to quantitative change rule in the traffic flow evolution process.Quantitative change is the prerequisite of qualitative change, and qualitative change is the accumulation of quantitative change, when quantitative change acquires a certain degree, will cause qualitative change, and then causes the change of traffic behavior.Quantitative change and qualitative change all belongs to traffic flow modes sudden change category, therefore, research traffic flow sudden change, not only to differentiate the qualitative change of traffic behavior, the also quantitative change of identification in time and height position thereof, could find in time that like this traffic behavior may occur abnormal, can also perform prevention work to the traffic abnormity that may occur in advance.For traffic flow sudden change and height identification problem thereof, traditional statistical requires traffic flow parameter to meet specific distribution, and need set up a large amount of pattern feature models, makes its practicality limited; The former grade of the Wang Xiao of Shandong Technology Univ is searched for height with average Analysis of Changing Points methods such as least square, minimum variances, although avoided the requirement to the traffic flow parameter distribution character, but be suitable for analyzing the basicly stable situation of traffic flow average, the situation adaptability that the traffic flow average is gradually changed is not good.The differentiation of traffic flow has very strong correlativity in time, show that the variation angle of traffic parameter on time orientation meets certain rule, and, at present to the research of traffic flow modes sudden change and height problem thereof, from traffic parameter, the variation angle on time orientation is not analyzed.
Summary of the invention
In view of this, the invention provides a kind of Traffic Flow Time Series height recognition methods based on vectorial similarity, can in the situation that traffic flow parameter gradually change in time, variation angle from traffic parameter on time orientation, the sudden change of judgement traffic flow modes and height position thereof, the traffic abnormity that prevention may occur in advance.
The present invention solves the problems of the technologies described above by following technological means:
Traffic Flow Time Series height recognition methods based on vectorial similarity, comprise the steps:
1) by the section traffic flow parameter in detecting device acquisition testing highway section, and store the traffic flow parameter database into;
2) read in chronological order traffic flow parameter from the traffic flow parameter database, set up the traffic direction quantity set of Traffic Flow Time Series;
3), according to the traffic direction quantity set, set up the vector similarity function of Traffic Flow Time Series, to obtain the similarity degree of each time point front and back traffic flow parameter development law;
4) identify the type of each point in Traffic Flow Time Series according to the duration of the similarity degree of traffic flow parameter development law before and after each time point and variation, then judge whether to undergo mutation, and identify the position that sudden change occurs.
Further, in described step 1), traffic flow parameter is at least one in the magnitude of traffic flow, speed and occupation rate.
Further, in described step 1), the acquisition time of traffic flow parameter is spaced apart GT, GT ∈ [3min, 8min].
Further, described step 2) in, the traffic flow parameter that the traffic direction quantity set of Traffic Flow Time Series is discrete arrangement in chronological sequence is linked in sequence formed vectorial the set:
TFV={TFV
j,i|j>i;i=1,2,…,n,j=1,2,…,n};
In above formula, the traffic direction quantity set that TFV is Traffic Flow Time Series, TFV
j,ibe carved into j change vector constantly while meaning traffic flow from i, its expression formula is as follows:
TFV
j,i=(Q
j-Q
i,j-i);
In above formula, Q
jwith Q
ibe respectively the j moment and i traffic flow parameter constantly.
Further, in described step 3), the vector similarity function of Traffic Flow Time Series obtains by following formula,
In above formula, function VS
i(j, k) means the vector similarity function of Traffic Flow Time Series, wherein, j>i>k, and j, i, k=1,2 ... n.
Further, described step 4), specifically comprise the steps:
Be carved into the i similarity of Changing Pattern constantly while being carved into i+1 moment Changing Pattern and k while 41) differentiating traffic flow from i, if similarity is greater than threshold value, put Q
ifor normal point, otherwise enter step 42);
Be carved into the i+1 similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and i while 42) differentiating traffic flow from i+1, if similarity is greater than threshold value, enter step 43), otherwise enter step 44);
43) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, and some Q
i+1for normal point, otherwise put Q
ifor pseudo-height;
Be carved into the i similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and k while 44) differentiating traffic flow from i, if similarity is greater than threshold value, put Q
ifor normal point, some Q
i+1for singular point, otherwise enter step 45);
45) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, some Q
i+1for pseudo-height, otherwise put Q
ifor pseudo-height;
If some Q
ifor height, illustrate that traffic flow undergos mutation at this some place, the position of the generation of height is i.
Traffic Flow Time Series height recognition methods based on vectorial similarity of the present invention, variation angle from traffic parameter on time orientation, based on vectorial similarity principle, whether the similarity degree changed before and after each point in developing by the identification traffic flow is differentiated differentiation and is occurred extremely, and comes the generation of identification sudden change and the position of height in conjunction with developing abnormal lasting duration.The method there is certain validity and sensitivity stronger, effective identification of the generation, particularly quantitative change suddenlyd change by the method identification in time traffic flow modes, can be used as the prompting of traffic flow anomaly trend, prevent in advance the generation of traffic qualitative change, and then implement to induce timely and effectively.
Embodiment
Fig. 1 shows the schematic flow sheet of the Traffic Flow Time Series height recognition methods based on vectorial similarity;
Fig. 2 shows the traffic flow parameter data that read in embodiment;
Fig. 3 shows the differentiation result of Traffic Flow Time Series mid point in embodiment;
Fig. 4 shows the differentiation schematic flow sheet for Traffic Flow Time Series point.
Embodiment
Below with reference to accompanying drawing, the present invention is described in detail.
Referring to Fig. 1, the Traffic Flow Time Series height recognition methods based on vectorial similarity, comprise the steps:
1) by the section traffic flow parameter in detecting device acquisition testing highway section, and store the traffic flow parameter database into; Described section traffic flow parameter can be the magnitude of traffic flow, speed, occupation rate etc., and this example is selected the magnitude of traffic flow; The acquisition time of traffic flow parameter is spaced apart GT, and the value of GT should be suitable, and too little Yi too much brings the random disturbance of traffic flow into, too conference affects the accuracy of recognition result, generally, within 3-8 minute, to be advisable, in the present embodiment, adopt the traffic flow parameter that the time interval is 5 minutes; This example of described detection highway section is selected certain Shang Mou highway section, rapid transit; Described traffic flow parameter be retrieved as prior art, no longer describe in detail;
2) read in chronological order traffic flow parameter from the traffic flow parameter database, set up the traffic direction quantity set (Traffic Flow Vector, TFV) of Traffic Flow Time Series; In the present embodiment, the data that the traffic flow parameter read is 07:30-12:30 one day, amount to 60 points, as shown in Figure 2;
Obtained traffic flow parameter in chronological sequence is linked in sequence, sets up the traffic direction quantity set (TFV) of Traffic Flow Time Series:
TFV={TFV
j,i|j>i;i=1,2,…,60,j=1,2,…,60}
Wherein, TFV
j,ibe carved into j change vector constantly while meaning traffic flow from i;
TFV
j,i=(Q
j-Q
i, j-i), ignore unit, constantly i and time t(t are the actual time after turning to minute, as 7:30 is 450) corresponding relation be
3), according to traffic direction quantity set (TFV), set up vector similarity (Vector Similarity, the VS) function of Traffic Flow Time Series, to obtain the similarity degree of each time point front and back traffic flow parameter development law; In described step 3), the vector similarity of Traffic Flow Time Series (VS) function adopts the included angle cosine measure, and its expression formula is shown below:
VS
i(j, k) is carved into the similarity of the Changing Pattern in the i moment while meaning when traffic flow is from i to be carved into j Changing Pattern constantly with k.Wherein, j > i > k, and j, i, k=1,2 ... n.VS
ithe value of (j, k) is larger, be carved into i Changing Pattern constantly while being carved into j Changing Pattern constantly to k when traffic flow is from i more similar, otherwise otherness is larger, for the possibility of height also just larger.
For distinguishing the similarity degree of each time point front and back traffic flow parameter development law of Traffic Flow Time Series, according to vector similarity threshold value (Vector Similarity Threshold, VST), the vector similarity value of quantification is divided into to two grades of height: if VS
i(j, k)>=VST, similarity is high; Otherwise similarity is low, as shown in the table:
4) identify the type of each point in Traffic Flow Time Series according to the duration of the similarity degree of traffic flow parameter development law before and after each time point and variation, and then whether the identification sudden change occurs, and identify the position of sudden change generation; The foundation that the Traffic Flow Time Series sudden change occurs is that exchange time sequence development law occurs that abnormal and abnormal lasting duration is not less than threshold value ST, in the present embodiment, and ST=10min;
In described Traffic Flow Time Series, the type of each point is for the identification height defines, and comprises improper point and normal point, and improper point comprises singular point and pseudo-height, and normal point comprises height and normal point.Wherein, singular point refers to carve at a time traffic parameter and occurs extremely, but does not change the point of traffic flow development law; Pseudo-height refers to the height that should not occur, previous moment be point that before and after height and this point, the Changing Pattern of traffic flow constantly similarity is low or and previous height between the variable height of putting; Height has been guided the Traffic Flow Time Series development law to change and has been changed and continued the point that duration is not less than ST; Normal point i.e. point except singular point, pseudo-height, height;
Referring to Fig. 4, in the recognition methods of described Traffic Flow Time Series height, for a Q
ithe determination methods of whether suddenling change specifically comprises the steps:
While 41) differentiating traffic flow from i, be carved into i+1 when Changing Pattern is with k constantly, be carved into i constantly the similarity of Changing Pattern (k be with i constantly be close to a normal point), if the similarity height is put Q
ifor normal point, otherwise enter step 42);
Be carved into the i+1 similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and i while 42) differentiating traffic flow from i+1, if the similarity height enters step 43), otherwise enter step 44);
43) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, and some Q
i+1for normal point, otherwise put Q
ifor pseudo-height;
Be carved into the i similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and k while 44) differentiating traffic flow from i, if the similarity height is put Q
ifor normal point, some Q
i+1for singular point, otherwise enter step 45);
45) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, some Q
i+1for pseudo-height, otherwise put Q
ifor pseudo-height.
If some Q
ifor height, illustrate that traffic flow undergos mutation at this some place, the position of the generation of height is i.
For example, Q
k=61, Q
i=74, Q
i+1=70, Q
i+2=72, and set VST=0.7, put Q
ithe differentiation process as follows:
According to step 41), VS
i(i+1, k)=-0.94868<VST, the similarity grade that is carved into i moment Changing Pattern while being carved into i+1 moment Changing Pattern and k while meaning traffic flow from i is low;
According to step 42), VS
i+1(i+2, i)=-0.75926<VST, the similarity grade that is carved into i+1 moment Changing Pattern while being carved into i+2 moment Changing Pattern and i while meaning traffic flow from i+1 is low;
According to step 44), VS
i(i+2, k)=-0.65079<VST, the similarity grade that is carved into i moment Changing Pattern while being carved into i+2 moment Changing Pattern and k while meaning traffic flow from i is low;
According to step 45), some Q
ithe lasting duration of an its immediate upper height is 15min>10min, put Q
ifor height, and some Q
i+1for pseudo-height;
The old friend is through-flow at a Q
iplace undergos mutation, and the moment that sudden change occurs is t=5i+445.
The differentiation result of described embodiment Traffic Flow Time Series mid point as shown in Figure 3.As shown in Figure 3, the Traffic Flow Time Series height recognition methods based on vectorial similarity detects altogether 12 heights (solid triangle) in example, 8 singular points (filled circles), 4 pseudo-heights (hollow triangle).Reach first peak at 7:50 left and right flow in the stage in the morning, now be in morning peak period, subsequently basic held stationary.From 8:40, flow starts to continue to rise, until the 10:25 left and right starts again steady decline.
For check height recognition result, each section flow between height done to linear regression, obtain as a result shown in following table.
The linear regression statistics
As seen from the above table, between each height, the Linear Quasi of flow is right better, and α<0.05 is arranged.Simultaneously, before and after each height, the intercept of linear fit function is different with slope, has verified that height front and back traffic flow has sudden change to occur.Can find out, before and after height, the Changing Pattern of the magnitude of traffic flow has different parameter attributes, but before and after the part height, the magnitude of traffic flow changes not remarkable, be mainly that its parameter attribute is remarkable not as qualitative change due to these traffic flow modes generation quantitative changes constantly, result shows that the method susceptibility is stronger.
Finally explanation is, above embodiment is only unrestricted in order to technical scheme of the present invention to be described, although with reference to preferred embodiment, the present invention is had been described in detail, those of ordinary skill in the art is to be understood that, can modify or be equal to replacement technical scheme of the present invention, and not breaking away from aim and the scope of technical solution of the present invention, it all should be encompassed in the middle of claim scope of the present invention.
Claims (6)
1. the Traffic Flow Time Series height recognition methods based on vectorial similarity, is characterized in that: comprise the steps:
1) by the section traffic flow parameter in detecting device acquisition testing highway section, and store the traffic flow parameter database into;
2) read in chronological order traffic flow parameter from the traffic flow parameter database, set up the traffic direction quantity set of Traffic Flow Time Series;
3), according to the traffic direction quantity set, set up the vector similarity function of Traffic Flow Time Series, to obtain the similarity degree of each time point front and back traffic flow parameter development law;
4) identify the type of each point in Traffic Flow Time Series according to the duration of the similarity degree of traffic flow parameter development law before and after each time point and variation, then judge whether to undergo mutation, and identify the position that sudden change occurs.
2. the Traffic Flow Time Series height recognition methods based on vectorial similarity as claimed in claim 1, it is characterized in that: in described step 1), traffic flow parameter is at least one in the magnitude of traffic flow, speed and occupation rate.
3. the Traffic Flow Time Series height recognition methods based on vectorial similarity as claimed in claim 1, it is characterized in that: in described step 1), the acquisition time of traffic flow parameter is spaced apart GT, GT ∈ [3min, 8min].
4. the Traffic Flow Time Series height recognition methods based on vectorial similarity as claimed in claim 1, it is characterized in that: described step 2), the traffic flow parameter that the traffic direction quantity set of Traffic Flow Time Series is discrete arrangement in chronological sequence is linked in sequence formed vectorial the set:
TFV={TFV
j,i|j>i;i=1,2,…,n,j=1,2,…,n};
In above formula, the traffic direction quantity set that TFV is Traffic Flow Time Series, TFV
j,ibe carved into j change vector constantly while meaning traffic flow from i, its expression formula is as follows:
TFV
j,i=(Q
j-Q
i,j-i);
In above formula, Q
jwith Q
ibe respectively the j moment and i traffic flow parameter constantly.
5. the Traffic Flow Time Series height recognition methods based on vectorial similarity as claimed in claim 1, it is characterized in that: in described step 3), the vector similarity function of Traffic Flow Time Series obtains by following formula,
In above formula, function VS
i(j, k) means the vector similarity function of Traffic Flow Time Series, wherein, j>i>k, and j, i, k=1,2 ... n.
6. the Traffic Flow Time Series height recognition methods based on vectorial similarity as described as any one in claim 1-5, it is characterized in that: described step 4) specifically comprises the steps:
Be carved into the i similarity of Changing Pattern constantly while being carved into i+1 moment Changing Pattern and k while 41) differentiating traffic flow from i, if similarity is greater than threshold value, put Q
ifor normal point, otherwise enter step 42);
Be carved into the i+1 similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and i while 42) differentiating traffic flow from i+1, if similarity is greater than threshold value, enter step 43), otherwise enter step 44);
43) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, and some Q
i+1for normal point, otherwise put Q
ifor pseudo-height;
Be carved into the i similarity of Changing Pattern constantly while being carved into i+2 moment Changing Pattern and k while 44) differentiating traffic flow from i, if similarity is greater than threshold value, put Q
ifor normal point, some Q
i+1for singular point, otherwise enter step 45);
45) differentiate some Q
ithe lasting duration of an its immediate upper height, if be more than or equal to ST, put Q
ifor height, some Q
i+1for pseudo-height, otherwise put Q
ifor pseudo-height;
If some Q
ifor height, illustrate that traffic flow undergos mutation at this some place, the position of the generation of height is i.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310406641.2A CN103488878B (en) | 2013-09-09 | 2013-09-09 | Traffic Flow Time Series height recognition methods based on vector similarity |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201310406641.2A CN103488878B (en) | 2013-09-09 | 2013-09-09 | Traffic Flow Time Series height recognition methods based on vector similarity |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103488878A true CN103488878A (en) | 2014-01-01 |
CN103488878B CN103488878B (en) | 2016-08-24 |
Family
ID=49829094
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201310406641.2A Active CN103488878B (en) | 2013-09-09 | 2013-09-09 | Traffic Flow Time Series height recognition methods based on vector similarity |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103488878B (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104318795A (en) * | 2014-10-31 | 2015-01-28 | 重庆大学 | Expressway site traffic state deviation degree acquiring method based on time-space analysis |
CN106506556A (en) * | 2016-12-29 | 2017-03-15 | 北京神州绿盟信息安全科技股份有限公司 | A kind of network flow abnormal detecting method and device |
CN108615361A (en) * | 2018-05-10 | 2018-10-02 | 江苏智通交通科技有限公司 | Crossing control time division methods and system based on multidimensional time-series segmentation |
CN110288003A (en) * | 2019-05-29 | 2019-09-27 | 北京师范大学 | Data variation recognition methods and equipment |
CN110889963A (en) * | 2018-09-11 | 2020-03-17 | 深圳云天励飞技术有限公司 | Road monitoring method, device and storage medium |
CN111125184A (en) * | 2019-11-23 | 2020-05-08 | 同济大学 | Bus passenger flow dynamic monitoring method based on time sequence structural variable point identification |
CN111613049A (en) * | 2019-02-26 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Road state monitoring method and device |
CN113362597A (en) * | 2021-06-03 | 2021-09-07 | 济南大学 | Traffic sequence data anomaly detection method and system based on non-parametric modeling |
CN114724390A (en) * | 2022-04-21 | 2022-07-08 | 浙江商汤科技开发有限公司 | Traffic signal control method and device, electronic device and storage medium |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
ES2354330A1 (en) * | 2009-04-23 | 2011-03-14 | Universitat Pompeu Fabra | Method for calculating measures of similarity between time signals |
CN102890862A (en) * | 2012-09-29 | 2013-01-23 | 浙江大学 | Traffic condition analyzing device and method based on vector mode |
CN103050005A (en) * | 2012-11-16 | 2013-04-17 | 北京交通大学 | Method and system for space and time analysis of urban road traffic states |
-
2013
- 2013-09-09 CN CN201310406641.2A patent/CN103488878B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
ES2354330A1 (en) * | 2009-04-23 | 2011-03-14 | Universitat Pompeu Fabra | Method for calculating measures of similarity between time signals |
CN102890862A (en) * | 2012-09-29 | 2013-01-23 | 浙江大学 | Traffic condition analyzing device and method based on vector mode |
CN103050005A (en) * | 2012-11-16 | 2013-04-17 | 北京交通大学 | Method and system for space and time analysis of urban road traffic states |
Non-Patent Citations (2)
Title |
---|
李强: "交通流时间序列的聚类分析方法及应用", 《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》 * |
王晓原,等: "交通流突变分析的变点统计方法研究", 《中国公路学报》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104318795A (en) * | 2014-10-31 | 2015-01-28 | 重庆大学 | Expressway site traffic state deviation degree acquiring method based on time-space analysis |
CN106506556A (en) * | 2016-12-29 | 2017-03-15 | 北京神州绿盟信息安全科技股份有限公司 | A kind of network flow abnormal detecting method and device |
CN106506556B (en) * | 2016-12-29 | 2019-11-19 | 北京神州绿盟信息安全科技股份有限公司 | A kind of network flow abnormal detecting method and device |
US11451566B2 (en) | 2016-12-29 | 2022-09-20 | NSFOCUS Information Technology Co., Ltd. | Network traffic anomaly detection method and apparatus |
CN108615361A (en) * | 2018-05-10 | 2018-10-02 | 江苏智通交通科技有限公司 | Crossing control time division methods and system based on multidimensional time-series segmentation |
CN110889963A (en) * | 2018-09-11 | 2020-03-17 | 深圳云天励飞技术有限公司 | Road monitoring method, device and storage medium |
CN110889963B (en) * | 2018-09-11 | 2021-07-20 | 深圳云天励飞技术有限公司 | Road monitoring method, device and storage medium |
CN111613049B (en) * | 2019-02-26 | 2022-07-12 | 北京嘀嘀无限科技发展有限公司 | Road state monitoring method and device |
CN111613049A (en) * | 2019-02-26 | 2020-09-01 | 北京嘀嘀无限科技发展有限公司 | Road state monitoring method and device |
CN110288003A (en) * | 2019-05-29 | 2019-09-27 | 北京师范大学 | Data variation recognition methods and equipment |
CN110288003B (en) * | 2019-05-29 | 2022-01-18 | 北京师范大学 | Data change identification method and equipment |
CN111125184A (en) * | 2019-11-23 | 2020-05-08 | 同济大学 | Bus passenger flow dynamic monitoring method based on time sequence structural variable point identification |
CN113362597A (en) * | 2021-06-03 | 2021-09-07 | 济南大学 | Traffic sequence data anomaly detection method and system based on non-parametric modeling |
CN114724390A (en) * | 2022-04-21 | 2022-07-08 | 浙江商汤科技开发有限公司 | Traffic signal control method and device, electronic device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN103488878B (en) | 2016-08-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103488878A (en) | Vector similarity based traffic flow time sequence change point identification method | |
CN103646534B (en) | A kind of road real-time traffic accident risk control method | |
WO2020244288A1 (en) | Method and apparatus for evaluating truck driving behaviour based on gps trajectory data | |
CN102360525B (en) | Discriminant analysis-based high road real-time traffic accident risk forecasting method | |
CN103646542B (en) | A kind of Forecasting Methodology of Evaluating traffic impact area and device | |
CN102829974B (en) | LMD (local mean decomposition) and PCA (principal component analysis) based rolling bearing state identification method | |
CN101581940B (en) | Tunnel event detection method based on integrated learning time sequence prediction | |
CN103198147B (en) | The differentiation of automatic monitoring abnormal data and processing method | |
CN106530127A (en) | Complaint early warning and monitoring analysis system based on text mining | |
CN109612729A (en) | A kind of malfunction recognition methods of port Machine Trolley Wheel Bearings | |
CN103473540A (en) | Vehicle track incremental modeling and on-line abnormity detection method of intelligent traffic system | |
CN102183620A (en) | Method for judging water source quality pollution and pollution level | |
CN104807640A (en) | Method of detecting life state of spatial rolling bearing based on vibration-sensitive time-frequency characteristics | |
CN100481153C (en) | Method for automatically inspecting highway traffic event based on offset minimum binary theory | |
CN111368642A (en) | Method for identifying modal frequency of railway ballastless track steel rail based on wheel-rail excitation | |
CN103473928A (en) | Urban traffic jam distinguishing method based on RFID technology | |
CN101271625A (en) | Method for detecting freeway traffic event by integration supporting vector machine | |
CN103927874A (en) | Automatic incident detection method based on under-sampling and used for unbalanced data set | |
CN108108890A (en) | A kind of defective workmanship risk analysis method based on process incidence relation | |
CN103278343A (en) | Method for identifying key structural fatigue component | |
WO2017107790A1 (en) | Method and apparatus for predicting road conditions using big data | |
CN103198707B (en) | A kind of vehicle regulate and control method based on traffic flow character dangerous under fine day situation | |
CN102592448A (en) | Method for testing safety performance of unsignalized intersection by utilizing equivalent traffic conflict | |
CN108961748A (en) | A kind of intersection operational safety state arrangement method based on the illegal rate of equivalent | |
CN109035763A (en) | Expressway traffic accident primary and secondary based on C4.5 is because of analysis and accident pattern judgment method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20201228 Address after: 402460 station No.14, no.6, 10th floor, innovation and development center, No.19 Lingfang Avenue, Changzhou street, Rongchang District, Chongqing Patentee after: Chongqing kezhiyuan Technology Co.,Ltd. Address before: 400030 No. 174 Sha Jie street, Shapingba District, Chongqing Patentee before: Chongqing University |
|
TR01 | Transfer of patent right |