CN104715604A

CN104715604A - Method and system for acquiring real-time traffic status information

Info

Publication number: CN104715604A
Application number: CN201410014230.3A
Authority: CN
Inventors: 郝勇刚; 朱江; 胡佳妮
Original assignee: Hangzhou Hikvision Digital Technology Co Ltd
Current assignee: Gaodewei Intelligent Traffic System Co., Ltd., Shanghai
Priority date: 2014-01-13
Filing date: 2014-01-13
Publication date: 2015-06-17
Anticipated expiration: 2034-01-13
Also published as: CN104715604B

Abstract

The invention relates to the field of intelligent transportation, and discloses a method and system for acquiring real-time traffic status information. The method includes the following steps that all road sections in a road network are traversed, so that effective links with the starting nodes and the terminal nodes being each provided with a checkpoint device are generated; the vehicle passing data of all the checkpoint devices in the current road condition period or the previous historical period within the license plate backtracking period are read in; the effective links are read one by one, and vehicle passing records corresponding to the starting nodes and the terminal nodes of the links are screened out in the vehicle passing data; the vehicle passing data matched with license plates are searched for, accordingly, the section vehicle speed and the reliability of single samples are calculated, and abnormal samples are filtered; the samples of all the effective links are distributed to all the road sections to generate the real-time traffic status information. According to the method and system, foundation road network data are processed to generate the effective links, the license recognition vehicle passing data are analyzed on this basis, the concept of the sample reliability is introduced, and the accuracy rate of the real-time traffic status information and the information coverage of the complex road network can be improved.

Description

Obtain method and the system thereof of real-time road condition information

Technical field

The present invention relates to intelligent transportation field, the method for the acquisition real-time road condition information particularly in traffic guidance and system thereof.

Background technology

Perfect along with correlation technique, significantly improving of car plate capture rate and discrimination, the traffic information collection technology based on Car label recognition data is arisen at the historic moment.Compare other traffic information collection technology, the traffic information collection technology based on license plate identification data has the advantages such as work continuity is strong, data accuracy is high, detection sample size is large.Based on the dynamic transport data gathering and processing system of license plate automatic discrimination technology by carrying out discriminatory analysis at two adjacent check points to the car plate of same car, the parameters such as the journey time of vehicle, travel speed can be obtained.If have many cars through specific road section in the given period, the average travel time in this section and average travel speed can also be obtained.

Publication number be 1937000 patent of invention disclose a kind of method that city intersection remote monitoring and vehicle flux recognition detect.In vehicle flux recognition detection method wherein, the oriented section in city road network between two signal lamps is considered as a processing unit by this invention.On-the-spot at traffic monitoring, by installing its video image of camera collection to this oriented section, in order to the needs transmitted carry out compression process to this image.Then the detection ring that virtual reality road surface is placed is set, digital denoising and image enhaucament is carried out to the image in virtual ring region, utilizes the virtual ring detection method of fuzzy Judgment to the vehicle Auto-counting by virtual detection ring.

Although this invention is a processing unit with the oriented section in city road network between two signal lamps, statistical study be the wagon flow data of intersection, the traffic information on section cannot be provided.

Publication number is a kind of method that the patent of invention of 102426783A discloses low discharge road traffic accident based on vehicle tracking and detects, this invention is by following the tracks of the vehicle of low discharge road, comparison vehicle license mates, eliminate the flase drop that vehicle identification causes, judge the generation of traffic events.The advantage of this invention to reduce the rate of false alarm that road traffic accident detects under traffic conditions, and the road traffic accident that can be applicable to low discharge road network or low discharge period detects, and is traffic hazard rescue Warning Service.

The shortcoming of this invention must meet low discharge owing to detecting section and close two conditions, and must lay the requirement of multiple high definition tollgate devices on this road different sections of highway, so the condition of data acquisition is comparatively strict, do not possess widely used condition in city.

Publication number be 101546478 patent of invention disclose a kind of based on the traffic flow integrated and distributed analytic system of license plate identification data and the method for process thereof, which region the traffic flow that this invention can be analyzed in the inherent special time period in appointed area derives from, and these traffic flows arrive one's respective area travel the path of process, and provide the quantized data and ratio that mainly collect pathway traffic stream; Meanwhile, the traffic flow analyzed in the inherent special time period in appointed area dissipates to arrive which land, and these traffic flows reach the driving path of these terminals, and provides quantized data and the ratio of main dissipation pathway traffic stream.

Although this invention, by identifying that car plate data obtain long distance vehicle running path, cannot calculate interval traffic information accurately, not possess the ability of middle microcosmic traffic information process.

Publication number is that the patent of invention of 103258430A discloses a kind of road trip time statistics and traffic decision method and device, this invention is captured in setting-up time section, sails section into and rolled away from the car plate data of compact car vehicle in section by the downstream road junction in section by the crossing, upstream in section; Car plate coupling is carried out to the crossing, upstream in section and the car plate data of downstream road junction collection; Determine that vehicle that car plate mates is by the running time in section and travel speed; After carrying out abnormal speed screening according to the travel speed determined, determine sample vehicle; According to the running time of each sample vehicle by section, in conjunction with the time period residing for current time, calculate statistical value hourage in section.

The shortcoming of this invention is the car plate data only gathering compact car vehicle in road trip time statistic processes, ignores the data of other vehicles, so sample size can be subject to certain impact; The filtration of exceptional sample only by anomaly sieving speed of a motor vehicle sample, and does not consider other factor, and this has considerable influence to the accuracy rate of traffic information; The scope of application of this invention is section, and is not suitable for the road network of city complexity.

In sum, currently available technology can only provide crossing or respective road segment real-time road, real-time road condition information accuracy rate is not high, applicable elements is comparatively strict.Need a kind of real-time road condition information that can either obtain high-accuracy, there is again the method and system of the acquisition real-time road condition information of higher urban complex road network coverage rate.

Summary of the invention

The object of the present invention is to provide a kind of method and the system thereof that obtain real-time road condition information, this invention generates active link by process basic electronic road net data in road network, board is analyzed on active link basis and knew car data, and introduce the concept of credibility of sample's, the real-time road condition information of higher accuracy can be obtained, effectively solve the real-time road condition information covering problem of urban complex road network simultaneously.

For solving the problems of the technologies described above, embodiments of the present invention disclose a kind of method obtaining real-time road condition information, and road network comprises section and node, and on part road-net node, coupling has the alert equipment of card, recorded car data, and method comprises the following steps:

Each section in traversal road network, have at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, generate active link, the start node of active link and terminal node are laid with card and warn equipment;

Read in the alert equipment of all cards of a upper history cycle in current road conditions cycle or car plate backtracking cycle car data;

Read active link one by one, filter out and the start node of the active link of current reading and terminal node corresponding car record respectively crossing in car data of the alert equipment of all cards;

Cross in car record with the start node of the active link of current reading and terminal node are corresponding respectively the mistake car record found car plate and mate what filter out, car record of crossing according to car plate coupling obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of single sample, reject with a low credibility in the exceptional sample of the first predetermined threshold;

After having traveled through all active links, the sample after each active link rejecting abnormalities sample is shared each section, each section generates real-time road condition information according to the sample shared separately.

Embodiments of the present invention also disclose a kind of system obtaining real-time road condition information, and road network comprises section and node, and on part road-net node, coupling has the alert equipment of card, recorded car data, and system comprises:

Active link generation module, for traveling through each section in road network, have at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, generate active link, the start node of active link and terminal node are laid with card and warn equipment;

Crossing car data receiver module, warning the car data excessively of equipment for reading in all cards of a upper history cycle in current road conditions cycle or car plate backtracking cycle;

Cross car record screening module, for after reading active link one by one, filter out and the start node of the active link of current reading and terminal node corresponding car record respectively crossing in car data of the alert equipment of all cards;

Car plate matching module, crosses in car record with the start node of the active link of current reading and terminal node are corresponding respectively the car record excessively found car plate and mate for what filter out in screening module;

Sample process module, obtains continuing through the start node of active link and the vehicle list sample of terminal node for the car record of crossing according to car plate coupling, calculates the confidence level of single sample, reject with a low credibility in the exceptional sample of the first predetermined threshold;

Multisample Fusion Module, for after having traveled through all active links, has shared each section by the sample after each active link rejecting abnormalities sample, and each section generates real-time road condition information according to the sample shared separately.

Compared with prior art, the key distinction and effect thereof are embodiment of the present invention:

This invention generates active link by process basic electronic road net data in road network, board is analyzed on active link basis and knew car data, and introduce the concept of credibility of sample's, the real-time road condition information of higher accuracy can be obtained, effectively solve the real-time road condition information covering problem of urban complex road network simultaneously.

Further, in the present invention, anomaly sieving speed of a motor vehicle sample is not only considered in the filtration of exceptional sample, and by introducing confidence level criterion, angularly consider from the vehicle attribute of vehicle sample, car plate attribute, travelling characteristic attribute, farthest ensure that the sample size needed for calculating, and high-quality sample accurately can be filtered out, by the real-time road condition information after multisample fusion calculation, there is higher data accuracy.

Accompanying drawing explanation

Fig. 1 is a kind of schematic flow sheet obtaining the method for real-time road condition information in first embodiment of the invention;

Fig. 2 is a kind of overall flow figure obtaining the method for real-time road condition information in first embodiment of the invention;

Fig. 3 a to Fig. 3 b is a kind of algorithm flow chart obtaining the method for real-time road condition information in first embodiment of the invention;

Fig. 4 a kind ofly in first embodiment of the invention obtains the process flow diagram generating active link in the method for real-time road condition information;

Fig. 5 is a kind of structural representation obtaining the system of real-time road condition information in third embodiment of the invention;

Fig. 6 is a kind of structural representation obtaining the system of real-time road condition information in four embodiment of the invention.

Embodiment

In the following description, many ins and outs are proposed in order to make reader understand the application better.But, persons of ordinary skill in the art may appreciate that even without these ins and outs with based on the many variations of following embodiment and amendment, also can realize each claim of the application technical scheme required for protection.

For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiments of the present invention are described in further detail.

First, following table 1 is depicted as some terminological interpretations related in the present invention:

Table 1

First embodiment of the invention relates to a kind of method obtaining real-time road condition information, Fig. 1 is the schematic flow sheet of the method for this acquisition real-time road condition information, in the method for this acquisition real-time road condition information, road network comprises section and node, on part road-net node, coupling has the alert equipment of card, recorded car data, specifically, as shown in Figure 1, the method for this acquisition real-time road condition information comprises the following steps:

In a step 101, each section in traversal road network, have at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, generate active link, the start node of active link and terminal node are laid with card and warn equipment." the alert equipment of card " refers to that electric police grasp shoot machine, bayonet socket video detector, wisdom monitor video detecting device etc. can carry out the equipment of Car license recognition detection.

Preferably, before step 101, further comprising the steps of:

Read basic road net data, basic road net data comprises the alert equipment of card and road-net node matched data;

Read configuration parameter, configuration parameter comprises the road conditions cycle, and road condition judgment threshold and history cross car data storing path; The road conditions cycle is the time interval that adjacent twice traffic information is issued, and road condition judgment threshold is for judging road traffic state, and traffic behavior comprises unimpeded, and slowly and block up, the road traffic state threshold value of different brackets is different;

According to basic road net data, continue before setting up the follow-up section topological sum in each independent section in basic road network section topological relation.

Follow-up section is using the terminal node of target road section as section start node, and by the initial transitable section, traffic flow direction of this node.

Before section of continuing be using the start node of target road section as section terminal node, and end at the transitable section, traffic flow direction of this node.

Preferably, before step 101, also step is comprised:

Whether be provided with the alert equipment of card according to the start node of target road section and terminal node, target road section is classified;

When only having start node or terminal node one place to be provided with the alert equipment of card, corresponding find target road section follow-up section or before to continue section, form active link with this target road section, stored in active link set;

When start node and terminal node are equipped with card police equipment, by this target road section directly stored in active link set.

Further, above-mentioned for when only having start node or terminal node one place to be provided with card police equipment, corresponding find target road section follow-up section or before to continue section, form in the step of active link with target road section, adopt the method for a step recursion, the method that a step recursion generates active link comprises following sub-step:

Corresponding searching only has start node place to be provided with the follow-up section of the target road section of the alert equipment of card or section of continuing before only having terminal node place to be provided with the target road section of the alert equipment of card, judges whether the terminal node in this follow-up section or the start node in front section of continuing have the alert device numbering of card;

If do not block alert device numbering, then continue to find, until arrive first crossing backward or forward; If there is the alert device numbering of card, then generate active link;

To continue before if also do not search out after arriving first crossing, terminal node has the follow-up section of the alert device numbering of card or start node to have card to warn device numbering section, read the right-hand rotation in section, current crossing, turn left and keep straight on section, stored in this follow-up section, the first crossing or before continue section set;

Read the first follow-up section, crossing or front section of continuing in the set of section one by one, judge whether the terminal node in the follow-up section of reading from this set or the start node in front section of continuing have the alert device numbering of card, if do not block alert device numbering, then continue to find, until arrive second crossing backward or forward; If there is the alert device numbering of card, then generate active link.

In addition, be appreciated that, one step recursion refers to when only having a place to be laid with card police equipment in target road section start node or terminal node, according to road topology relation, find in its follow-up section or front section of continuing and can form with target road section the section that two end nodes are laid with card police, will the two virtual group synthesis satisfactory effective road chain.Because only finding the section of first crossing of upstream or downstream appearance, so be called " a step recursion ".

In other embodiments of the present invention, be not limited to the mode of a step recursion, also can find the section of other crossings of upstream or downstream appearance, form active link with target road section.

After this enter step 102, read in the alert equipment of all cards of a upper history cycle in current road conditions cycle or car plate backtracking cycle car data.

In addition, be appreciated that, the car data of crossing that the alert equipment of the card that part road-net node is provided with records can be described as " car data known by board ", and it comprises the number-plate number, crosses the information such as car time, car speed, crossing serial number, lane number, direction, track, transport condition, type of vehicle.

The introducing in car plate backtracking cycle considers: in the ordinary course of things, program is by the car source data excessively in the road conditions cycle of process, generate real-time road condition information, and when under the situations such as section is long or section road conditions are blocked up, within a road conditions cycle, section cannot obtain the same car plate data by playing terminal note, so need forward trace in time, in units of road conditions period distances, in " car plate backtracking cycle " (such as 30 minutes), carry out searching of same car plate data.

After this enter step 103, read the active link generated by step 101 one by one, the crossing in car data of the alert equipment of all cards read in a step 102 filters out and the start node of the active link of current reading and terminal node corresponding car record respectively.

Preferably, in step 103, following sub-step is comprised:

That reads the alert equipment of all cards one by one crosses car data;

Judge whether the current car data excessively read in belongs to the current road conditions cycle;

If the current road conditions cycle, first the crossing serial number this being crossed car data compares with the alert device numbering of the card at the terminal node place of the active link of current reading, then compares with the alert device numbering of the card at the start node place of the active link of current reading; If not the current road conditions cycle, only this is crossed the crossing serial number of car data and compare with the alert device numbering of the card at the start node place of the active link of current reading;

If the alert device numbering of the card crossing the crossing serial number of car data and active link terminal node place is identical and to cross the entrance driveway direction of car data identical with the termination direction, section of active link, then this is crossed car data and cross car set of records ends stored in this active link terminal node, judge whether to travel through all car datas excessively;

If the alert device numbering of the card crossing the crossing serial number of car data and active link start node place is identical, then this is crossed car data and cross car set of records ends stored in this active link start node, judge whether to travel through and allly cross car data;

If the crossing serial number crossing car data is not identical with the alert device numbering of card at active link terminal node or start node place, directly judge whether to travel through all car datas excessively;

If do not traveled through all car datas excessively, then read next and crossed car data.

In addition, be appreciated that, because read in be in current road conditions cycle or car plate backtracking cycle history cycle cross car data, there is not the car data excessively in future time, if so the car data excessively read in is the current road conditions cycle, first the crossing serial number this being crossed car data compares with the alert device numbering of the card at the terminal node place of the active link of current reading, then compares with the alert device numbering of the card at start node place.If not the current road conditions cycle, namely the car data of crossing read in is that a cycle interior upper history cycle recalled by car plate, only this is crossed the crossing serial number of car data to compare with the alert device numbering of the card at the start node place of the active link of current reading, just can search out vehicle and continue through the start node of active link and the car record excessively of terminal node.

After this step 104 is entered, cross in car record with the start node of the active link of current reading and terminal node are corresponding respectively the mistake car record found car plate and mate what filter out, car record of crossing according to car plate coupling obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of single sample, reject with a low credibility in the exceptional sample of the first predetermined threshold.

In addition be appreciated that, in other embodiments of the present invention, rejecting abnormalities sample can also be realized abnormal data denoising by interquartile range method, and is not limited to the method for the lower exceptional sample of above-mentioned rejecting confidence level.

Preferably, at step 104, following sub-step is comprised:

Read active link terminal node one by one and cross car record, active link start node cross in car record find that the car plate of crossing car record with the active link terminal node of current reading matches cross car record;

If do not search out the car plate of coupling, next of reading active link terminal node crosses car record;

If search out the car plate of coupling, judge the situation whether having car plate repeated matching in matching process.If do not repeat, car record of crossing according to car plate coupling obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of vehicle list sample, and filter the exceptional sample of credibility of sample's lower than the first predetermined threshold, next of reading active link terminal node crosses car record;

If there is repetition, got the car time large cross car record after calculate overall speed and the confidence level of single sample again, and filter with a low credibility in the exceptional sample of the first predetermined threshold, next reading active link terminal node crosses car record.

After this enter step 105, after having traveled through all active links, the sample after each active link rejecting abnormalities sample is shared each section, each section generates real-time road condition information according to the sample shared separately.

Preferably, in step 105, following sub-step is comprised:

According to the overall speed of road condition judgment threshold and sample, add up the sample size dropped between clear area;

When the ratio of the sample size in when between clear area and total sample size is greater than or equal to the 3rd predetermined threshold, take credibility of sample's as weight, the overall speed weighted mean of each single sample between clear area is generated real-time road condition information;

When the ratio of the sample size in when between clear area and total sample size is less than the 3rd predetermined threshold, take credibility of sample's as weight, all single sample interval speed of a motor vehicle weighted means are generated real-time road condition information.

After this this flow process is terminated.

Second embodiment of the invention relates to a kind of method obtaining real-time road condition information, and the second embodiment improves on the basis of the first embodiment, and main improvements are:

First, after the step 102 of reading in all car datas excessively blocking the equipment of police of a history cycle in current road conditions cycle or car plate backtracking cycle, step is comprised:

Pre-service is carried out to retain the effective car data excessively correctly recognizing car plate to the car data excessively read in.

Secondly, after obtaining continuing through the start node of active link and the vehicle list sample of terminal node at step 104, calculate in the sub-step of the confidence level of vehicle list sample, calculate in conjunction with the attribute relevant to type of vehicle and vehicle running characteristics, association attributes comprises from high to low successively according to shared weight during the single credibility of sample's of calculating:

Start node entrance driveway Vehicular turn attribute, attribute comprises through vehicles, right-turning vehicles, left turning vehicle, head end operation, and the weight shared by above-mentioned attribute arranges from high to low.

Terminal node entrance driveway vehicle lane attribute, attribute comprises Through Lane, right-turn lane, left turn lane, and the weight shared by above-mentioned attribute arranges from high to low.

Vehicle attribute, attribute comprises contour of the vehicle size and type of vehicle, and the weight shared by contour of the vehicle is greater than the weight shared by type of vehicle.

Car plate attribute, attribute comprises local licence plate and nonlocal licence plate, and the weight shared by local licence plate is greater than the weight shared by nonlocal licence plate.

In addition, be appreciated that in other embodiments of the present invention, also can determine according to other modes the weight that each attribute is shared when calculating single credibility of sample's, and be not limited to the mode defined in present embodiment.

Finally, the sample after each active link rejecting abnormalities sample is being shared each section, each section comprises step before generating the step 105 of real-time road condition information according to the multiple samples shared separately:

Judge whether the ratio of the sample size that car plate mates and total sample size reaches the second predetermined threshold,

If do not reach the second predetermined threshold, then that reads a car plate backtracking cycle interior upper history cycle crosses car data to increase sample size.

In present embodiment, anomaly sieving speed of a motor vehicle sample is not only considered in the filtration of exceptional sample, and by introducing confidence level criterion, angularly considers, can filter out high-quality sample accurately from the vehicle attribute of vehicle sample, car plate attribute, travelling characteristic attribute; Meanwhile, farthest ensure that the sample size needed for calculating by ratio shared by judgement sample amount, by the real-time road condition information after multisample fusion calculation, there is higher data accuracy.

As the preference of present embodiment, Fig. 2 is to Figure 4 shows that overall flow figure, algorithm flow chart and active link product process figure.

Design philosophy mainly considers that car plate is the unique identity card of automobile, electricity police or the tollgate devices of crossing is stopped by being laid in section initial sum, the license plate of the complete vehicle by section can be obtained, inquire about the time of this vehicle by two continuous crossing detecting devices, mistiming can be obtained, combining road length, then can calculate the bicycle overall speed of this vehicle by target road section.When there is multiple single sample in target road section, must merge multiple sample, and then the interval average speed of target road section can be obtained.

The sample vehicle in target approach section is owing to may come from each entrance driveway direction, crossing, upstream, and the direction in crossing inlet road, downstream is well-determined when rolling target road section away from, so can by the numbering of upstream and downstream crossing and downstream intersection entrance driveway direction, target road section and mistake car data are mated, by license plate, the car plate that upstream and downstream crossing is caught is mated, calculate the overall speed of single sample, and calculate the confidence level of this sample according to the state priori such as type and travelling characteristic of single sample.

Specifically, as shown in Figure 2, board knowledge real-time road overall flow figure is made up of 12 functional modules shown in following table 2:

Table 2

Above-mentioned functions module generates active link according to basic road net data, after car plate coupling is carried out to mistake car data in the basis of active link, processes and shares each section, generate real-time road condition information to sample.Therefore, overall flow is divided into two major parts, is respectively to process basic road net data and mistake car data, and thus, the input of whole flow process comprises:

1) mistake car data: the car data excessively that electricity is warned and bayonet socket collects being laid in crossing, section

Table 3

2) basic road net data: the alert equipment of basic road property data, card and road-net node matched data

Table 4

3) configuration parameter: road conditions cycle, each grade road traffic state threshold value

	Megapolis	Large size city	Medium-sized city	Micropolis	District
						Through street	20-40	20-40	20-40	20-40	20-40
Main road	12-25	12-25	12-25	12-25	12-25
						Main line	10-23	10-23	10-23	10-23	10-23
Branch road	8-20	8-20	8-20	8-20	8-20

Note: its state threshold of different cities scale should not be identical, tentative same units: Km/h

Table 5

Consider that exporting vehicle speed data is convenient to the utilizations such as traffic forecast from now on, induction, signal control, so real-time road condition information returns with speed of a motor vehicle form.

The traffic information (unit: km/h) exported

	Road section ID	Category of roads	The forward speed of a motor vehicle	The reverse speed of a motor vehicle
					Road example drawn by two-wire	10001	4	45	-1
Road example drawn by single line	10002	4	45	22

Table 6

Fig. 3 a, 3b and Figure 4 shows that algorithm flow chart and active link product process figure.As shown in Fig. 3 a, 3b, algorithm flow comprises the steps:

1, initialization:

1) basic road net data is read: comprise basic road net data and block alert equipment and road-net node matched data;

2) read configuration parameter: cycle time parameter, each grade road conditions state threshold and history cross car data storing path;

2, set up road network topology: according to line drawing road single, double in electronic chart, respectively in formation base road network each independent section follow-up section topological sum before to continue section topological relation.

3, section in basic road net data set is read.

4, section topological classification classification:

Whether mate according to target road section upstream and downstream section and have the alert equipment of card, the topological relation in section can be divided into four classes:

Class1: have initial sum to stop crossing numbering

Type 2: have and stop crossing numbering

Type 3: have initial crossing to number

Type 4: stop crossing numbering without initial sum

5, carry out section " a step recursion " by topological classification and generate active link:

In reality, because the license plate identification equipment construction cost compare based on high definition is high, so can exist because cloth sets up an office, position is not enough causes the problem that Detection Information spatial coverage is not high.Devise for this situation in algorithm design and build effective road chain and cross " a step recursion " in car, namely when only having a place to be laid with card police equipment in target road section upstream and downstream crossing, according to road topology relation, find in its follow-up section or front section of continuing and can form with target road section the section that upstream and downstream crossing is laid with card police, the synthesis of the two virtual group is met effective road chain of algorithm requirement, then carry out car plate coupling.Consider the accuracy of real-time road, the path culculating problem temporarily travelled by this vehicle, the solution in the follow-up or front section of continuing that is designed to only to derive, i.e. " a step recursion ".

Class1: do not carry out " a step recursion " directly stored in active link set;

Type 2: section of continuing before finding current independent section, enters " generating active link module ";

Type 3: the follow-up section finding current independent section, enters " generating active link module ";

Type 4: do not carry out " a step recursion ", directly enter step 3;

6, generate active link, idiographic flow as shown in Figure 4:

1) finding topological classification is section (finding the follow-up section that topological classification is the target road section of 3) of continuing before the target road section of 2, whether the quantity in section (topological classification is the follow-up section of 3) of continuing before judgement is greater than 0, if be greater than 0, then enter step 2), otherwise exit and generate active link module, flow process terminates;

2) judge whether to reach crossing, be, the section set that continues before finding (topological classification is the follow-up section set of 3) enters step 4), otherwise enters step 3);

3) initial plant numbering (topological classification be 3 judged whether that termination type is numbered) is judged whether, there is numbering then by current road segment and front section of continuing (topological classification be 3 be then follow-up section) virtual merging, generate active link, flow process terminates, otherwise returns step 1);

4) to continue before reading successively every bar independence section in section set (topological classification is the follow-up section set of 3), and judge initial plant numbering (topological classification be 3 judged whether that termination type is numbered), there is numbering then by current road segment and front section of continuing (topological classification be 3 be then follow-up section) virtual merging, generate active link and enter step 7), otherwise enter step 5);

5) finding topological classification is section (finding the follow-up section that topological classification is the target road section of 3) of continuing before the target road section of 2, whether the quantity in section (topological classification is the follow-up section of 3) of continuing before judgement is greater than 0, if be greater than 0, then enter step 6), otherwise enter step 7);

6) judge whether to reach crossing, be, enter step 7), otherwise enter step 4);

7) last record is gathered in section (topological classification is the follow-up section of 3) of continuing before judging whether to reach, and is that flow process terminates, otherwise returns step 4).

7, judging whether to generate active link, is the active link of generation is put into effective link set, otherwise returns step 3;

8, judge whether to reach road network table end record, be enter step 9, otherwise return step 3;

9, read in the car data (at present for 5 minutes) excessively in Tn cycle, if Tn=To, then the car data excessively read in is the car data excessively in current period;

10, road section ID in active link set is read one by one.

11, car data pre-service excessively: retaining transport condition is normal car data excessively;

12, the car data excessively in car data set was read one by one;

13, whether the current cycle of reading in is current real-time period, is enter step 14, otherwise enters step 15;

14, CrossLSH(crossing serial number is judged) whether number identical with termination crossing, section, be enter step 16, otherwise enter step 15;

15, CrossLSH(crossing serial number is judged) whether number identical with initial crossing, section, be then this is crossed car record stored in VehiclePass_F data acquisition, enter step 17;

16, DirectIndex(entrance driveway direction is judged) whether identical with termination direction, section, be then this is crossed car record stored in VehiclePass_T data acquisition, enter step 17;

17, judge whether to arrive car record sheet end, be enter step 18, otherwise return step 12;

18, car plate coupling: find coupling according to the car plate data in VehiclePass_T in VehiclePass_F, get the record that time value is large when car plate repeats, otherwise enter step 19;

19, the process of single sample matches car plate:

1) single sample interval speed of a motor vehicle calculates: overall speed=road section length/(PassTime2-PassTime1) * 3.6

2) single credibility of sample's calculates (in conjunction with type of vehicle and vehicle running characteristics):

A) initial crossing inlet road Vehicular turn attribute: right-turning vehicles, through vehicles, left turning vehicle, head end operation;

B) crossing inlet road vehicle lane attribute is stopped: right-turn lane, Through Lane, left turn lane;

C) vehicle attribute: 1) profile: cart, middle car, dolly

2) character: common car, special car;

D) car plate possession attribute: local licence plate, nonlocal licence plate.

In 4 attribute that above single credibility of sample's calculates, the shared weight of initial crossing inlet road Vehicular turn property value (hereafter representing with a) is the highest, next stops crossing inlet to vehicle lane attribute (hereafter representing with b), being vehicle attribute (hereafter representing with c) again, is finally car plate possession attribute (hereafter representing with d).That is, the magnitude relationship of weighted value is: a>b>c>d;

Below the weight magnitude relationship of each feature in every attribute:

A) in: craspedodrome > right-hand rotation > left-hand rotation > reverses end for end

B) in: craspedodrome > right-hand rotation > turns left

C) in: profile > character: car > cart in dolly > in profile;

The special car of common car > in character

D) in: the licence plate of local licence plate > other places

A typical example is: have two samples, the sample attribute of X and Y, X is described as: a keeps straight on, and b keeps straight on, c dolly, the common car of c, the local licence plate of d; The sample attribute of Y is described as: a keeps straight on, and b turns left, c cart, the common car of c, d other places licence plate.Through the calculating of confidence level modular algorithm, the confidence level (0.9) of sample X is higher than sample Y confidence level (0.5).

20, exceptional sample is rejected:

1) the too low single sample (travelling characteristic method) of confidence level is rejected

2) abnormal data denoising (interquartile range method)

21, judge that in the Tn cycle, whether car plate all mates end, is enter step 22, otherwise returns step 18;

22, judge whether to arrive active link set end record, be enter step 23, otherwise return step 10;

23, judge whether to need the upper cycle history read in the car plate backtracking cycle to cross car data, be enter step 24, otherwise enter step 25.Rule of judgment is that (such as 3%) does not need to read history and crosses car data when the ratio of the sample size and the total sample of termination direction, section entrance driveway that match car plate is more than or equal to x%, otherwise needs to read history and cross car data and increase sample size;

24, judge whether the whole car datas excessively read in the car plate backtracking cycle, be enter step 25, otherwise return step 9;

25, sample splits each independent section:

Active link is split as each independent section, and complete single sample by active link is shared each independent section;

26, the section in being gathered by the independent section formed after active link fractionation is read one by one;

27, the multisample speed of a motor vehicle merges:

According to road conditions threshold value (unimpeded, slow, block up), add up the sample size dropped between clear area, when unimpeded sample size is more than or equal to Y% with the ratio of total sample size (such as 10%), then the single sample interval speed of a motor vehicle weighted mean (credibility of sample's is weight) between clear area calculated, result exports as the result after the multisample fusion in this independent section; When unimpeded sample size is less than Y% with the ratio of total sample size, be then weighted on average by all single samples, result exports as result after fusion;

28 judge whether that arriving active link splits rear independent section set footline record, is enter step 29, otherwise returns step 26;

29, road network traffic information generates, and returns platform;

30, flow process terminates.

Each method embodiment of the present invention all can realize in modes such as software, hardware, firmwares.No matter the present invention realizes with software, hardware or firmware mode, instruction code can be stored in the addressable storer of computing machine of any type (such as permanent or revisable, volatibility or non-volatile, solid-state or non-solid, fixing or removable medium etc.).Equally, storer can be such as programmable logic array (Programmable Array Logic, be called for short " PAL "), random access memory (Random Access Memory, be called for short " RAM "), programmable read only memory (Programmable Read Only Memory, be called for short " PROM "), ROM (read-only memory) (Read-Only Memory, be called for short " ROM "), Electrically Erasable Read Only Memory (Electrically Erasable Programmable ROM, be called for short " EEPROM "), disk, CD, digital versatile disc (Digital Versatile Disc, be called for short " DVD ") etc.

Third embodiment of the invention relates to a kind of system obtaining real-time road condition information, and Fig. 5 is the structural representation of the system of this acquisition real-time road condition information.

Specifically, the system road network of this acquisition real-time road condition information comprises section and node, and on part road-net node, coupling has the alert equipment of card, recorded car data, and as shown in Figure 5, this system comprises:

Preferably, the system obtaining real-time road condition information also comprises with lower module:

Initialization module, for reading basic road net data and configuration parameter.Basis road net data comprises the alert equipment of card and road-net node matched data, and configuration parameter comprises the road conditions cycle, and road condition judgment threshold and history cross car data storing path;

Section topology sets up module, and for according to basic road net data, continue before setting up the follow-up section topological sum in each independent section in basic road network section topological relation.

In addition, be appreciated that the road conditions cycle is the time interval that adjacent twice traffic information is issued.

Road condition judgment threshold is for judging road traffic state, and the road traffic state threshold value of different brackets is different.

Preferably, the system obtaining real-time road condition information also comprises:

Topological classification sort module, for whether being provided with the alert equipment of card according to the start node of target road section and terminal node, classifies to target road section;

When only having start node or terminal node one place to be provided with the alert equipment of card, active link generation module correspondence find target road section follow-up section or before to continue section, form active link with target road section, stored in active link set;

When start node and terminal node are equipped with the alert equipment of card, active link generation module by this target road section directly stored in active link set.

Preferably, crossing in car record screening module, comprising following submodule:

Crossing car data reading submodule, warning the car data excessively of equipment for reading all cards one by one;

The road conditions cycle judges submodule, for judging whether the current car data excessively read in belongs to the current road conditions cycle;

Numbering comparison sub-module, for when the road conditions cycle judges that submodule is judged as the current road conditions cycle, first the crossing serial number this being crossed car data compares with the alert device numbering of the card at the terminal node place of the active link of current reading, then compares with the alert device numbering of the card at the start node place of the active link of current reading.And during for judging that submodule is judged as not being the current road conditions cycle when the road conditions cycle, only this being crossed the crossing serial number of car data and comparing with the alert device numbering of the card at the start node place of the active link of current reading;

If the alert device numbering of the card crossing the crossing serial number of car data and active link terminal node place is identical and to cross the entrance driveway direction of car data identical with the termination direction, section of active link, then this is crossed car data and cross car set of records ends stored in this active link terminal node;

If the crossing serial number crossing car data is identical with the alert device numbering of the card at active link start node place, then this is crossed car data and cross car set of records ends stored in this active link start node.

Cross car data traversal submodule, cross car data for what judge whether to travel through the alert equipment of all cards, do not traveled through if be judged as and allly crossed car data, then cross car data reading submodule and read next and cross car data.

In addition, be appreciated that, because read in be in current road conditions cycle or car plate backtracking cycle history cycle cross car data, there is not the car data excessively in future time, if so current road conditions cycle, first the crossing serial number this being crossed car data compares with the alert device numbering of the card at the terminal node place of the active link of current reading, then compares with the alert device numbering of the card at terminal node place.If not the current road conditions cycle, only this is crossed the crossing serial number of car data to compare with the alert device numbering of the card at the start node place of the active link of current reading, just can search out vehicle and continue through the start node of active link and the car record excessively of terminal node.

Preferably, in car plate matching module, comprise following submodule:

Reading submodule, crosses car record for reading active link terminal node one by one;

Submodule found by coupling car plate, for crossing car record according to the active link terminal node read, active link start node cross in car record find car plate coupling cross car record;

If the car plate that submodule does not search out coupling found by coupling car plate, next of reading submodule reading active link terminal node crosses car record;

Car plate repeats to judge submodule, if find for coupling car plate the car plate that submodule searches out coupling, judges the situation whether having car plate repeated matching in matching process;

If do not repeat, sample process module obtains continuing through the start node of active link and the vehicle list sample of terminal node according to the car record of crossing of car plate coupling, calculate overall speed and the confidence level of vehicle list sample, and filter the exceptional sample of credibility of sample's lower than the first predetermined threshold, next of reading submodule reading active link terminal node crosses car record;

If there is repetition, got the car time large cross car record after, sample process module calculates overall speed and the confidence level of single sample again, and filters with a low credibility in the exceptional sample of the first predetermined threshold, and next of reading submodule reading active link terminal node crosses car record.

Preferably, sample process module comprises following submodule:

Overall speed calculating sub module, for the vehicle list sample computation interval speed of a motor vehicle of the start node continuing through active link that mates car plate and terminal node;

Credibility of sample's calculating sub module, for calculating credibility of sample's in conjunction with type of vehicle and vehicle operation characteristic;

Exceptional sample filters submodule, with a low credibility in the exceptional sample of the first predetermined threshold for filtering.

Preferably, multisample Fusion Module comprises following submodule:

Statistics submodule, for the overall speed according to road condition judgment threshold and sample, adds up the sample size dropped between clear area;

Fractional sample weighting submodule, when ratio for the sample size in when between clear area and total sample size is greater than or equal to the 3rd predetermined threshold, take credibility of sample's as weight, using the single sample interval speed of a motor vehicle weighted mean between clear area as multisample fusion results;

Overall situation sample weighting submodule, when the ratio for the sample size in when between clear area and total sample size is less than the 3rd predetermined threshold, take credibility of sample's as weight, using all single sample interval speed of a motor vehicle weighted means as multisample fusion results.

In the present invention, anomaly sieving speed of a motor vehicle sample is not only considered in the filtration of exceptional sample, and by introducing confidence level criterion, angularly consider from the vehicle attribute of vehicle sample, car plate attribute, travelling characteristic attribute, farthest ensure that the sample size needed for calculating, and high-quality sample accurately can be filtered out, by the real-time road condition information after multisample fusion calculation, there is higher data accuracy.First embodiment is the method embodiment corresponding with present embodiment, and present embodiment can be worked in coordination with the first embodiment and be implemented.The relevant technical details mentioned in first embodiment is still effective in the present embodiment, in order to reduce repetition, repeats no more here.Correspondingly, the relevant technical details mentioned in present embodiment also can be applicable in the first embodiment.

Four embodiment of the invention relates to a kind of system obtaining real-time road condition information, and the 4th embodiment improves on the basis of the 3rd embodiment, and main improvements are:

First, the system of this acquisition real-time road condition information also comprises:

Pretreatment module, for carrying out pre-service to retain the effective car data excessively correctly recognizing car plate to the car data excessively read in.

Secondly, cross car data receiver module and also comprise submodule:

History is crossed car data reading and is judged submodule, for judging whether the ratio of the sample size that car plate mates and total sample size reaches the second predetermined threshold, if do not reach the second predetermined threshold, then the history crossing a upper history cycle in the car data receiver module reading car plate backtracking cycle crosses car data to increase sample size.

As preference, system construction drawing as shown in Figure 6, the input of whole system comprised car data, basic road net data and other supplemental characteristics, cross car data read module and road network topology MBM respectively to crossing car data and basic road net data processes, core processing module is inputted after process, this core processing module generate active link go forward side by side driving board coupling, finally generate traffic information, crossing in car of generation traffic information, history road condition data also can be utilized to fill up traffic information.

Second embodiment is the method embodiment corresponding with present embodiment, and present embodiment can be worked in coordination with the second embodiment and be implemented.The relevant technical details mentioned in second embodiment is still effective in the present embodiment, in order to reduce repetition, repeats no more here.Correspondingly, the relevant technical details mentioned in present embodiment also can be applicable in the second embodiment.

The present invention leads over prior art in the road network coverage rate of real-time road condition information and the accuracy rate two of traffic information:

1, prior art mainly with intersection or independently section be processing unit, the spatial coverage of traffic information is very low, and the present invention is then take city road network as processing unit, can obtain higher spatial coverage.

2, prior art is comparatively strict to the requirement detecting section, must meet the requirement closed, not possess widely used condition in city.The present invention is then lower to the requirement detecting section, the requirement that demand fulfillment is not closed.

Although 3 prior aries, by identifying that car plate data obtain long distance vehicle running path, cannot calculate interval traffic information accurately.And the present invention not only possesses the ability that macroscopical vehicle running path differentiates, and there is the function of middle microcosmic real-time road condition information process.

4, prior art only gathers the car plate data of compact car vehicle in road trip time computation process, ignores the data of other vehicles, so sample size can be subject to certain impact; The filtration of exceptional sample only by anomaly sieving speed of a motor vehicle sample, and does not consider other factor, and this has considerable influence to the accuracy rate of traffic information.And the present invention is by introducing confidence level criterion, angularly consider from the vehicle attribute of vehicle sample, car plate attribute, travelling characteristic attribute, farthest ensure that the sample size needed for calculating, and high-quality sample accurately can be filtered out, by the real-time road condition information after multisample blending algorithm calculates, there is higher data accuracy.

It should be noted that, the each unit mentioned in the present invention's each equipment embodiment is all logical block, physically, a logical block can be a physical location, also can be a part for a physical location, can also realize with the combination of multiple physical location, the Physical realization of these logical blocks itself is not most important, and the combination of the function that these logical blocks realize is only the key solving technical matters proposed by the invention.In addition, in order to outstanding innovative part of the present invention, the unit not too close with solving technical matters relation proposed by the invention is not introduced by the above-mentioned each equipment embodiment of the present invention, and this does not show that the said equipment embodiment does not exist other unit.

It should be noted that, in the claim and instructions of this patent, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element " being comprised " limited by statement, and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.

Although by referring to some of the preferred embodiment of the invention, to invention has been diagram and describing, but those of ordinary skill in the art should be understood that and can do various change to it in the form and details, and without departing from the spirit and scope of the present invention.

Claims

1. obtain a method for real-time road condition information, it is characterized in that, road network comprises section and node, and on part road-net node, coupling has the alert equipment of card, and recorded car data, said method comprising the steps of:

Each section in traversal road network, have at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, generate active link, the start node of described active link and terminal node are laid with card and warn equipment;

Read active link one by one, filter out and the start node of the active link of current reading and terminal node corresponding car record respectively crossing in car data of the alert equipment of described all cards;

Cross in car record the mistake car record found car plate and mate the start node of described active link that is that filter out and current reading and terminal node are corresponding respectively, the car record of crossing mated according to described car plate obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of single sample, reject with a low credibility in the exceptional sample of the first predetermined threshold;

2. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, each section in described traversal road network having at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, before generating the step of active link, further comprising the steps of:

Read basic road net data, described basic road net data comprises the alert equipment of card and road-net node matched data;

Read configuration parameter, described configuration parameter comprises the road conditions cycle, and road condition judgment threshold and history cross car data storing path;

3. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, each section in described traversal road network having at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, before generating the step of active link, also comprise step:

When only having start node or terminal node one place to be provided with the alert equipment of card, corresponding find target road section follow-up section or before to continue section, generate active link with this target road section, stored in active link set;

4. the method for acquisition real-time road condition information according to claim 3, it is characterized in that, described when only having start node or terminal node one place to be provided with card police equipment, corresponding find target road section follow-up section or before to continue section, form in the step of active link with this target road section, adopt the method for a step recursion, the method for a described step recursion comprises following sub-step:

Corresponding find only have start node or terminal node place be provided with the target road section of the alert equipment of card follow-up section or before to continue section, judge whether the terminal node in this follow-up section or the start node in front section of continuing have card to warn device numbering;

Read described first follow-up section, crossing or front section of continuing in the set of section one by one, judge whether the terminal node in the follow-up section of reading from this set or the start node in front section of continuing have the alert device numbering of card, if do not block alert device numbering, then continue to find, until arrive second crossing backward or forward; If there is the alert device numbering of card, then generate active link.

5. the method for acquisition real-time road condition information according to claim 1, is characterized in that, described read in the alert equipment of all cards of a history cycle in current road conditions cycle or car plate backtracking cycle cross the step of car data after, comprise step:

Pre-service is carried out to retain the effective car data excessively correctly recognizing car plate to the described car data excessively read in.

6. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, active link is read one by one described, filter out cross in the step of car record corresponding with the start node of the active link of current reading and terminal node difference crossing in car data of the alert equipment of described all cards, comprise following sub-step:

That reads the alert equipment of described all cards one by one crosses car data;

7. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, cross in car record the mistake car record found car plate and mate the start node of described active link that is that filter out and current reading and terminal node are corresponding respectively, the car record of crossing mated according to described car plate obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of single sample, reject in the step of the exceptional sample in the first predetermined threshold with a low credibility, comprise following sub-step:

Read active link terminal node one by one and cross car record, cross in car record at active link start node and find coupling;

If search out the car plate of coupling, judge the situation whether having car plate repeated matching in matching process; If do not repeat, the car record of crossing mated according to described car plate obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of vehicle list sample, and filter the exceptional sample of credibility of sample's lower than the first predetermined threshold, next of reading active link terminal node crosses car record;

8. the method for acquisition real-time road condition information according to claim 7, it is characterized in that, in the sub-step of the confidence level of described calculating vehicle list sample, calculate in conjunction with the attribute relevant to type of vehicle and vehicle running characteristics, described attribute comprises from high to low successively according to shared weight during the single credibility of sample's of calculating:

Start node entrance driveway Vehicular turn attribute, described attribute comprises through vehicles, right-turning vehicles, left turning vehicle, head end operation, and the weight shared by above-mentioned attribute arranges from high to low;

Terminal node entrance driveway vehicle lane attribute, described attribute comprises Through Lane, right-turn lane, left turn lane, and the weight shared by above-mentioned attribute arranges from high to low;

Vehicle attribute, described attribute comprises contour of the vehicle size and type of vehicle, and the weight shared by contour of the vehicle is greater than the weight shared by type of vehicle;

Car plate attribute, described attribute comprises local licence plate and nonlocal licence plate, and the weight shared by local licence plate is greater than the weight shared by nonlocal licence plate.

9. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, described, sample after each active link rejecting abnormalities sample is shared each section, each section comprises step before generating the step of real-time road condition information according to the multiple samples shared separately:

10. the method for acquisition real-time road condition information according to claim 1, it is characterized in that, described, sample after each active link rejecting abnormalities sample is shared each section, each section generates in the step of real-time road condition information according to the sample shared separately, comprises following sub-step:

When the ratio of the sample size in when between clear area and total sample size is greater than or equal to the 3rd predetermined threshold, take credibility of sample's as weight, the single sample interval speed of a motor vehicle average weighted between clear area is generated real-time road condition information;

11. 1 kinds of systems obtaining real-time road condition information, it is characterized in that, road network comprises section and node, and on part road-net node, coupling has card to warn equipment, recorded car data, and described system comprises:

Active link generation module, for traveling through each section in road network, have at least a place to be provided with the situation of the alert equipment of card for the start node in section and terminal node, generate active link, the start node of described active link and terminal node are laid with card and warn equipment;

Cross car record screening module, for after reading active link one by one, filter out and the start node of the active link of current reading and terminal node corresponding car record respectively crossing in car data of the alert equipment of described all cards;

Car plate matching module, crosses in car record with the start node of the active link of current reading and terminal node are corresponding respectively the mistake car record found car plate and mate for what filter out in described screening module;

Sample process module, obtains continuing through the start node of active link and the vehicle list sample of terminal node for the car record of crossing mated according to described car plate, calculates the confidence level of single sample, reject with a low credibility in the exceptional sample of the first predetermined threshold;

The system of 12. acquisition real-time road condition informations according to claim 11, is characterized in that, also comprise with lower module:

Initialization module, for reading basic road net data and configuration parameter; Described basic road net data comprises the alert equipment of card and road-net node matched data, and described configuration parameter comprises the road conditions cycle, and road condition judgment threshold and history cross car data storing path;

The system of 13. acquisition real-time road condition informations according to claim 11, is characterized in that, also comprise:

When only having start node or terminal node one place to be provided with the alert equipment of card, described active link generation module correspondence find target road section follow-up section or before to continue section, form active link with target road section, stored in active link set;

When start node and terminal node are equipped with the alert equipment of card, described active link generation module by this target road section directly stored in active link set.

The system of 14. acquisition real-time road condition informations according to claim 11, is characterized in that, also comprise:

Pretreatment module, for carrying out pre-service to retain the effective car data excessively correctly recognizing car plate to the described car data excessively read in.

The system of 15. acquisition real-time road condition informations according to claim 11, is characterized in that, crosses in car record screening module, comprise following submodule described:

Crossing car data reading submodule, warning the car data excessively of equipment for reading described all cards one by one;

Numbering comparison sub-module, for when the road conditions cycle judges that submodule is judged as the current road conditions cycle, first the crossing serial number this being crossed car data compares with the alert device numbering of the card at the terminal node place of the active link of current reading, then compares with the alert device numbering of the card at the start node place of the active link of current reading; And during for judging that submodule is judged as not being the current road conditions cycle when the road conditions cycle, only this being crossed the crossing serial number of car data and comparing with the alert device numbering of the card at the start node place of the active link of current reading;

If the crossing serial number crossing car data is identical with the alert device numbering of the card at active link start node place, then this is crossed car data and cross car set of records ends stored in this active link start node;

Cross car data traversal submodule, cross car data for what judge whether to travel through the alert equipment of all cards, do not traveled through allly cross car data if be judged as, then described car data reading submodule of crossing reads next and crosses car data.

The system of 16. acquisition real-time road condition informations according to claim 11, is characterized in that, in described car plate matching module, comprises following submodule:

If the car plate that submodule does not search out coupling found by described coupling car plate, next of reading submodule reading active link terminal node crosses car record;

Car plate repeats to judge submodule, if find for described coupling car plate the car plate that submodule searches out coupling, judges the situation whether having car plate repeated matching in matching process;

If do not repeat, the car record of crossing that sample process module is mated according to described car plate obtains continuing through the start node of active link and the vehicle list sample of terminal node, calculate overall speed and the confidence level of vehicle list sample, and filter the exceptional sample of credibility of sample's lower than the first predetermined threshold, next of reading submodule reading active link terminal node crosses car record;

The system of 17. acquisition real-time road condition informations according to claim 11, is characterized in that, described sample process module comprises following submodule:

Overall speed calculating sub module, for the vehicle list sample computation interval speed of a motor vehicle of the start node continuing through active link that mates described car plate and terminal node;

The system of 18. acquisition real-time road condition informations according to claim 11, is characterized in that, described car data receiver module excessively, also comprises submodule:

The system of 19. acquisition real-time road condition informations according to claim 11, is characterized in that, described multisample Fusion Module comprises following submodule: