CN103150900A - Traffic jam event automatic detecting method based on videos - Google Patents

Abstract

The invention discloses a traffic jam event automatic detecting method based on videos. The method includes steps that (1) real-time traffic parameter information of monitoring points is obtained based on a video detecting device, the parameter information is transmitted to a background server to store in a real-time mode; (2) obliterated data are identified, abnormal data are filtered, and data normalization processing is implemented by using a z-score method; (3) historical data are extracted to conduct clustering analysis by using a automatic detecting processing device, road traffic state foundation clustering centers are generated by utilizing a clustering analysis method; (4) euclidean distances to each clustering center are calculated according to the real-time traffic information, a present traffic jam event is automatically judged, and processed data are released through a release terminal device; (5) the clustering centers of each traffic state are recalculated, iteration is repeated, and traffic jam event automatic detection is achieved. The traffic jam event automatic detecting method based on the videos is mainly used for collecting traffic information, reduces cost of traffic information collection and improves veracity and response speed of the traffic jam event automatic detecting.

Description

A kind of traffic congestion event automatic detection method based on video

Technical field

The present invention relates to traffic information collection and traffic behavior perception field, can reduce the human cost of traffic information collection, improve accuracy and reaction efficiency that the traffic congestion event detects automatically.

Background technology

the road traffic running status can be described with different traffic parameters, video detecting device can utilize the image and the image background that detect to carry out differential variation relatively, obtain the device that enters and sail out of information of vehicle target, thereby obtain the information such as vehicle passes through, existence, vehicle commander, the speed of a motor vehicle, adding up protocol analysis by the cycle can the multiple traffic parameter of Real-time Obtaining, various parameter informations return background server and store by the network real-time Transmission, but in existing traffic events for the using method of video detecting device mostly based on single traffic parameter, as the magnitude of traffic flow, speed, density etc., a plurality of parametric synthesis are not arranged to detect and carry out comprehensive analysis processing, but the traffic behavior dynamic change is extremely complicated, be difficult to reflect with single parameter its characteristic, single parameter setting can be caused the inaccuracy of differentiation, and be all by manually carrying out video monitoring when using video detecting device at present, waste of manpower and efficient are lower.

traffic is differentiated the traffic flow operation of the accuracy affects city road network of algorithm, directly a plurality of traffic parameters being set on the present video detecting device that uses can cause system to the erroneous judgement of state, the differentiation of mistake may cause the road section traffic volume load, form and stagnate, condition discrimination algorithm commonly used, as the California algorithm, exponential smoothing, standard deviation etc., all the too stiff variation setting threshold according to single traffic parameter is classified traffic behavior, the urban road traffic flow amount is extremely complicated, simple being described by means of certain traffic parameter can't obtain accurate condition discrimination, easily impact detects the degree of accuracy of the magnitude of traffic flow, the processing that affects emergency event causes the waste of public administration resource.

Summary of the invention

The object of the present invention is to provide a kind of human cost that can reduce traffic information collection, improve the automatic accuracy that detects of traffic congestion event and the traffic congestion event automatic detection method based on video of reaction efficiency.

For achieving the above object, the present invention has adopted following technical scheme: a kind of traffic congestion event automatic detection method based on video, the equipment that uses in the method comprises video detecting device, data communications equipment, the background server that is used for data storing and filtration, automatic Check processing equipment and issue terminal equipment, signal connection in order between described each equipment, it is characterized in that: the method comprises following step:

(1) based on video detecting device, obtain the real-time traffic parameter information of road to be measured chain check point, described traffic parameter comprises flow, occupation rate, car speed, time headway data, and described parameter information returns background server and stores through described data communications equipment real-time Transmission;

(2) obliterated data is identified, and abnormal data is carried out cleaning and filtering, utilize z-score to send out and carry out the data normalization processing;

(3) utilize described automatic Check processing equipment to carry out cluster analysis from the historical data that background server extracted month, adopt clustering methodology and generate that road is very unimpeded, unimpeded, four the traffic behaviors bases cluster centres of walking or drive slowly and block up;

(4) according to real-time transport information, calculate the Euclidean distance with each cluster centre, relatively choose the shortest cluster centre, then automatically judge the traffic congestion situation of current time road to be measured chain, at last the data of processing are released news through issue terminal equipment;

(5) recomputate the cluster centre of all kinds of traffic behaviors, wait for next data calculating, iterate, realize that the traffic congestion event detects automatically;

Described k-means clustering method model construction based on traffic parameter is to carry out computational analysis to all sample datas of flow, occupation rate, car speed, every kind of parameter of time headway, comprises the following steps:

(1) all historical data samples of one month are divided into k initial classes, then with the center of gravity (average) of this k class as initial mean vectors;

(2) all data samplers except congealing point are sorted out one by one, adopted Euclidean distance method, each sample is included into congealing point from its that nearest class, such congealing point is updated to the present average of this class, until all samples has all been returned class.

(3), repeating step (2), until till all samples all can not reallocate.

beneficial effect of the present invention: the present invention utilizes the video monitoring apparatus on road traffic to be detected highway section, the vehicle operating traffic parameter information of Real-time Obtaining road chain check point, comprise flow, occupation rate, car speed, time headway data, by Internet Transmission to background data base and store, obliterated data and abnormal data are identified and processed, the historical data of extracting one month is carried out data normalization and is processed, adopt the disposal route of k-means cluster analysis, generate very unimpeded, unimpeded, the four class traffic behavior classification centers of walking or drive slowly and block up, the data that the Real-time Obtaining video detector is passed back, calculate the Euclidean distance with each cluster centre, be judged to be current traffic behavior with the cluster centre of bee-line, generate new cluster centre, wait for the calculating of next data, so iterate, analyze and determine the traffic congestion of described road to be measured chain and unimpeded, realize that the block up automatic judgement of event of road traffic detects, utilized fully traffic flow, to flow, occupation rate, car speed, the time headway supplemental characteristic carries out overall treatment and judges, improved the accuracy that the traffic congestion event detects automatically, time and financial cost by artificial video monitoring traffic congestion have been reduced, increased the speed that emergency event reports and manages, can improve service efficiency and the service level of urban road.

Description of drawings

Fig. 1 is workflow diagram of the present invention;

System equipment connection diagram used in Fig. 2 Fig. 1;

Fig. 3 cluster analysis original sample distribution plan;

Design sketch after Fig. 4 k-means cluster analysis;

Fig. 5 is magnitude of traffic flow schematic diagram in 24 hours;

In figure 1, video detecting device, 2, data communications equipment, 3, background server,

4, automatic Check processing equipment, 5, issue terminal equipment.

Embodiment

as illustrated in fig. 1 and 2 a kind of based on video traffic congestion event automatic detection method workflow and the method in the equipment that uses, the equipment that uses comprises video detecting device 1, data communications equipment 2, the background server 3 that is used for data storing and filtration, automatic Check processing equipment 4 and issue terminal equipment 5, data communications equipment 2 and video detecting device 1 link together by cable on the video monitoring erecting frame that then is arranged on road, by data communications equipment 2 and background server 3, automatically between Check processing equipment 4 and issue terminal equipment 5 in order signal connects and carries out signal and transmit, based on comprise following several treatment step with the method after last equipment connection: (1) is based on the video detecting device 1 in video monitoring equipment, set the real-time traffic parameter flow of four parameter acquiring road to be measured chain check points in video detecting device 1, occupation rate, car speed, the time headway data, again the parameter information real-time Transmission is returned background server 3 and stored, (2) identify by 3 pairs of obliterated datas of background server, and abnormal data is carried out cleaning and filtering, enter automatic Check processing equipment 4 after filtration, a z-score Standardization Act that can utilize by the algorithm compiling is installed in automatic Check processing equipment 4 is carried out the software program that data normalization is processed, (3) extract the historical data of month, employing clustering methodology calculating road is very unimpeded, unimpeded, the cluster centre of four traffic behaviors of walking or drive slowly and block up, (4) according to real-time transport information, calculate the Euclidean distance with each cluster centre, then the traffic congestion situation of automatically judging current time road to be measured chain enters issue terminal equipment 5, and to carry out issue automatically open, (5) recomputate the cluster centre of all kinds of traffic behaviors, wait for next data calculating, execution in step four again, iterate, and realize that the traffic congestion event detects automatically.

As shown in Figure 3 because the differentiation traffic behavior underlying parameter unit of selecting is different, and the order of magnitude is different, do not have a comparability, overall treatment simultaneously is as judging quota, need to carry out standardization processing to data source, by functional transformation, its numerical value is mapped to certain numerical value interval, the z-score Standardization Act is data to be carried out the typical way of normalized, and the z-score standardized method is based on the average of raw data and standard deviation and carries out standardization.

Suppose that sample data integrates the (q as S={ ₁, o ₁, v ₁, g ₁), (q ₂, o ₂, v ₂, g ₂) ..., (q _n, o _n, v _n, g _n), (q wherein _i, o _i, v _i, g _i) sample point of expression, q _i, o _i, v _i, g _iRepresent respectively flow, occupation rate, speed and time headway, unit is respectively/hour, 1, km/hour and second.

The average of flow $\stackrel{\‾}{q}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{q}_i/n,$ The standard deviation of flow ${\mathrm{var}}_q=\sqrt{\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(q_i-\stackrel{\‾}{q})}^2};$

The average of occupation rate $\stackrel{\‾}{o}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{o}_i/n,$ The standard deviation of occupation rate ${\mathrm{var}}_o=\sqrt{\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(o_i-\stackrel{\‾}{o})}^2};$

The average of speed $\stackrel{\‾}{v}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{v}_i/n,$ The standard deviation of speed ${\mathrm{var}}_v=\sqrt{\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(v_i-\stackrel{\‾}{v})}^2};$

The average of time headway $\stackrel{\‾}{g}=\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{g}_i/n,$ The standard deviation of time headway ${\mathrm{var}}_g=\sqrt{\frac{1}{n}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}{(g_i-\stackrel{\‾}{g})}^2}.$

Input parameter form after normalization is $(\hat{q}=\frac{q_i-\stackrel{\‾}{q}}{{\mathrm{var}}_q},\hat{o}=\frac{o_i-\stackrel{\‾}{o}}{{\mathrm{var}}_o},\hat{v}=\frac{v_i-\stackrel{\‾}{v}}{{\mathrm{var}}_v},\hat{g}=\frac{g_i-\stackrel{\‾}{g}}{{\mathrm{var}}_g})$

The k-means clustering method of traffic behavior synthetic determination is one of data mining mode of commonly using, n sample set is divided into k bunch of (k＜n), allow sample condense to congealing point by certain principle, congealing point is constantly revised or iteration, until the sample standard deviation variance summation of each bunch inside reach given threshold value or iteration stable till, finally obtain the center (bunch center can be a virtual point) of each bunch, its objective function is as follows:

$J=\underset{j=1}{\overset{k}{\mathrm{\Σ}}}\underset{i=1}{\overset{n}{\mathrm{\Σ}}}\left|\right|x_i^{\left(j\right)}-c_j{\left|\right|}^2$

Wherein:

The sum of the single class sample set of n--;

Bunch number that the single class sample of k--is divided (k＜n);

The desired value of j--cluster analysis;

--sample point;

c _j--cluster centre;

--the distance metric between sample point and cluster centre;

For several video detectors as separate individuality, each individuality is chosen the historical data of month, data are divided into two groups: one group is the data source of regular working day, one group is data sources festivals or holidays, then respectively for setting up working day and festivals or holidays corresponding cluster centre matrix, formulate working day and festivals or holidays two kinds of different traffic discrimination standards.K-means clustering method based on traffic parameter builds model, very unimpeded according to the traffic behavior that will obtain, unimpeded, jogging and congestion level, all sample datas to flow, occupation rate, car speed, every kind of parameter of time headway are calculated respectively, comprise three steps:

(1) all historical data samples of one month are divided into 4 initial classes, then with the center of gravity (average) of these 4 classes as initial mean vectors;

(2) all data samplers except congealing point are sorted out one by one, adopted Euclidean distance method, each sample is included into congealing point from its that nearest class, such congealing point is updated to the present average of this class, until all samples has all been returned class.

(3) repeating step (2) is until till all samples all can not reallocate.

At first extract the historical data of month and make sample, utilize the z-score method to make the standard normalized to it, use the k-means cluster analysis, draw in the cluster of workaday four traffic behavior grades, working day, the center matrix of four traffic behavior cluster centres was as follows:

$M=(\hat{q},\hat{o},\hat{v},\tilde{g})=\begin{array}{cccc}-1.23& 0.78& -0.86& 0.57\\ 0.04& 0.47& -0.18& -0.23\\ -0.79& -0.96& -0.58& 0.36\\ 2.121& -2.69& 3.14& -1.95\end{array}$

Obtain the cluster centre point of each bunch by final mask, four cluster centres are respectively $({\tilde{q}}_1,{\tilde{o}}_1,{\tilde{v}}_1,{\tilde{g}}_1),({\tilde{q}}_2,{\tilde{o}}_2,{\tilde{v}}_2,{\tilde{g}}_2),({\tilde{q}}_3,{\tilde{o}}_3,{\tilde{v}}_3,{\tilde{g}}_3),({\tilde{q}}_4,{\tilde{o}}_4,{\tilde{v}}_4,{\tilde{g}}_4).$

According to real-time transport information, gather a certain moment to obtain one group of real time data (q, o, v, g), obtain through data z-score normalized Flow, occupation rate, car speed, time headway are calculated Euclidean distance with each cluster centre successively, and each is very unimpeded, unimpeded, jogging and congestion status cluster centre are $({\tilde{q}}_1,{\tilde{o}}_1,{\tilde{v}}_1,{\tilde{g}}_1),({\tilde{q}}_2,{\tilde{o}}_2,{\tilde{v}}_2,{\tilde{g}}_2),({\tilde{q}}_3,{\tilde{o}}_3,{\tilde{v}}_3,{\tilde{g}}_3),$

D1, d2, d3, d4 represent respectively Euclidean distance separately, and cluster centre condition judgement corresponding to minor increment is current time traffic congestion situation.

D1, d2, d3, d4 computing formula are distinguished as follows:

$\mathrm{<figure-callout\; id="1"\; label="d"\; filenames="130204181510.png,130204181523.png"\; state="\{\{state\}\}">d</figure-callout>}1=\sqrt{{(q-{\tilde{q}}_1)}^2+{(o-{\tilde{o}}_1)}^2+{(v-{\tilde{v}}_1)}^2+{(g-{\tilde{g}}_1)}^2)}$

$\mathrm{<figure-callout\; id="2"\; label="d"\; filenames="130204181510.png,130204181523.png"\; state="\{\{state\}\}">d</figure-callout>}2=\sqrt{{(q-{\tilde{q}}_2)}^2+{(o-{\tilde{o}}_2)}^2+{(v-{\tilde{v}}_3)}^2+{(g-{\tilde{g}}_4)}^2)}$

$\mathrm{<figure-callout\; id="3"\; label="d"\; filenames="130204181510.png,130204181523.png"\; state="\{\{state\}\}">d</figure-callout>}3=\sqrt{{(q-{\tilde{q}}_3)}^2+{(o-{\tilde{o}}_3)}^2+{(v-{\tilde{v}}_3)}^2+{(g-{\tilde{g}}_4)}^2)}$

$\mathrm{<figure-callout\; id="4"\; label="d"\; filenames="130204181523.png"\; state="\{\{state\}\}">d</figure-callout>}4=\sqrt{{(q-{\tilde{q}}_4)}^2+{(o-{\tilde{o}}_4)}^2+{(v-{\tilde{v}}_4)}^2+{(g-{\tilde{g}}_4)}^2)}$

Recomputate the cluster centre of all kinds of traffic behaviors, wait for the reception of next real time data, calculate the Euclidean distance of new data, upgrade traffic behavior, again calculate new cluster centre, so iterate, analyze and determine the traffic congestion of described road to be measured chain and unimpeded, realize that the block up automatic judgement of event of road traffic detects, be as shown in Figure 4 the design sketch after the k-means cluster analysis.

Be built in automatic Check processing equipment 4 by above-mentioned algorithm composing software program, working procedure calculates the Euclidean distance of each real-time parameter and cluster centre, and regenerates new cluster centre, carries out the data instance checking.The traffic behavior value of 24 hours is carried out check analysis to select as shown in Figure 5 on September 7,6:00 to 2010 year on the 6th September in 2010 between 6:00 from database, and data break 45 minutes amounts to 288 traffic behavior values, the time response of traffic behavior.

The present invention has utilized traffic flow fully, flow, occupation rate, car speed, time headway supplemental characteristic are carried out overall treatment to be judged, can improve the accuracy that the traffic congestion event detects automatically, reduce time and financial cost by artificial video monitoring traffic congestion, increase the speed that emergency event reports and manages, can improve service efficiency and the service level of urban road.

Claims (3)

1. traffic congestion event automatic detection method based on video, the equipment that uses in the method comprises video detecting device, data communications equipment, the background server that is used for data storing and filtration, automatic Check processing equipment and issue terminal equipment, signal connection in order between described each equipment, it is characterized in that: the method comprises following step:

(1) based on video detecting device, obtain the real-time traffic parameter information of road to be measured chain check point, described traffic parameter comprises flow, occupation rate, car speed, time headway data, and described parameter information returns background server and stores through described data communications equipment real-time Transmission;

(2) obliterated data is identified, and abnormal data is carried out cleaning and filtering, utilize z-score to send out and carry out the data normalization processing;

(3) utilize described automatic Check processing equipment to carry out cluster analysis to the historical data of extracting, adopt clustering methodology to calculate and generate that road is very unimpeded, unimpeded, the basic cluster centre of four traffic behaviors of walking or drive slowly and block up;

(4) according to real-time transport information, calculate the Euclidean distance with each cluster centre, the shortest cluster centre is chosen in contrast, then automatically judges the traffic congestion situation of current time road to be measured chain, at last the data of processing is released news through issue terminal equipment;

(5) recomputate the cluster centre of all kinds of traffic behaviors, wait for next data calculating, iterate, realize that the traffic congestion event detects automatically.

2. the traffic congestion event automatic detection method based on video according to claim 1 is characterized in that: the described historical data of extracting from background server is one month.

3. the traffic congestion event automatic detection method based on video according to claim 2, it is characterized in that: described k-means clustering method model construction based on traffic parameter is to carry out computational analysis to all sample datas of flow, occupation rate, car speed, every kind of parameter of time headway, comprises the following steps:

(1) all the historical data samples in month are divided into k initial classes, then with the center of gravity of this k class as initial mean vectors;

(2) all data samplers except congealing point are sorted out one by one, adopted Euclidean distance method, each sample is included into congealing point from its that nearest class, such congealing point is updated to the present average of this class, until all samples has all been returned class;

(3) repeating step (2) is until till all samples all can not reallocate.

Images