CN106991812A - The detection method and device of row array formula crossing array area vehicle flowrate based on video mode - Google Patents
The detection method and device of row array formula crossing array area vehicle flowrate based on video mode Download PDFInfo
- Publication number
- CN106991812A CN106991812A CN201710313072.5A CN201710313072A CN106991812A CN 106991812 A CN106991812 A CN 106991812A CN 201710313072 A CN201710313072 A CN 201710313072A CN 106991812 A CN106991812 A CN 106991812A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- virtual coil
- coil identifier
- covering
- identifier
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
- G08G1/0116—Measuring and analyzing of parameters relative to traffic conditions based on the source of data from roadside infrastructure, e.g. beacons
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a kind of detection method and device of the row array formula crossing array area vehicle flowrate based on video mode, method includes：A background image) is extracted；B present image) is obtained, and it is matched with background image；C) judge whether matching value is more than first threshold, in this way, be determined as the covering of no vehicle, do not cover that count is incremented, perform D)；Otherwise, it is determined that to there is vehicle covering, count is incremented for covering, and counting is not covered and is set to 0, A is returned to)；D) judge do not cover whether counting is more than the first setting frame number or matching value is more than Second Threshold, in this way, perform E)；Otherwise, A is returned)；E) judge that covering is counted and whether be more than the second setting frame number, in this way, vehicle count increase performs F)；Otherwise, F is performed)；F) covering counting is set to 0, return to step A).The present invention can realize quick effective detection vehicle and count the magnitude of traffic flow, need not destroy original pavement structure, reduce secondary input and maintenance cost.
Description
Technical field
The present invention relates to intellectual traffic control field, more particularly to a kind of row array formula crossing array area car based on video mode
The detection method and device of flow.
Background technology
With the sharp increase of city vehicle number, traffic pressure becomes increasing, and jam occurs in city
Probability is increasing.Urban congestion problem is increasingly becoming the subject matter of traffic control department processing.For urban transportation, congestion is existing
As if it is extremely fatal.The time resource that jam does not only result in related personnel is greatly wasted, and can also cause environmental resource
Severe contamination.It can be said that congestion directly affects social productive forces.
Find by inquiry, under urban environment, the reason for triggering traffic congestion mainly there are three kinds：
1st, occurring abnormality on road causes lane capacity to decline, such as car accident, road construction etc..This
The congestion of type, needs traffic police to carry out in-situ processing after finding in time and dredges.
2nd, urbanization process aggravation causes existing crossing design unreasonable.In conventional road construction, there is road network system
Function is indefinite, the problems such as hierarchical organization is unreasonable.Region has lacked the secondary distributor road and branch road of collecting and distributing effect, does not form suitable
Road network.Primary and secondary in road network system, branch road network it is out of proportion.Branch road to the shunting action of trunk roads traffic not
Substantially.Such congestion is, it is necessary to carry out the coordination between land use and Transport Facility Construction.The traffic administration system in city
System is supporting with urban planning, city construction system, the need for meeting Traffic Development.
3rd, crossing major-minor arterial highway traffic light time is improper.Traditional traffic grade mode is difficult to again scientifically command car
It is current, efficiently solve traffic blocking problem.For example when North and South direction vehicle flowrate is very big, traditional control method can not
Ensure that North and South direction green light opens the longer time, the problem of to alleviate traffic congestion.Simultaneously can not be by the green light of east-west direction
Time control causes traffic capacity to waste in the reasonable time.Such traffic congestion is, it is necessary to obtain crossing all directions
The magnitude of traffic flow, reasonably control signal lamp is to realize that rational science is shunted.
Publication No. CN 1441369A, it is entitled《Traffic control method and facility》Application for a patent for invention in, propose
The method of road traffic control and device facility adaptable therewith, indicate that control motor vehicle, non-motor vehicle are each for guiding
To or it is some to all or part of vehicle shunting.It according to default or automatic sensing optimising and adjustment and the program step of intelligence,
Wagon flow is arranged in oldered array, realizes multilane without the wagon flow intersected or the intersected less side that continuously or discontinuously row array is passed through
Method (abbreviation row array formula is passed through), can maximize and improve the traffic efficiency at crossing.
Either common crossing scheme or row array formula transit scenario, red green according to the current direction of the control of real-time traffic flow amount
Lamp, can effectively alleviate the problem of non-rice habitats abnormality causes congestion.The main method of existing vehicle detection is as follows：
It is entitled in Publication No. CN 104464294A《A kind of road section traffic volume state evaluation based on array radar
Method》Application for a patent for invention in propose the method based on radar.The mounted array radar vehicle detection device by section to be measured,
And distinguish detection zone and blind area.The real-time traffic parameter that array radar vehicle equipment will obtain the detection zone in section to be measured is believed
Breath.This method blind area is larger, does not apply to control traffic lights in real time.
It is CN 104268589A in patent publication No., it is entitled《A kind of front vehicles detection method》Invention it is special
The front vehicles detection method based on machine learning is proposed in profit application.Positive negative example base is set up, collection is a large amount of with the presence of vehicle
With the video pictures without vehicle, trained the image for having vehicle in video pictures as vehicle in positive sample, Picking video picture
There is no the picture formation negative example base of vehicle.Vehicle sections feature extraction in image comes out hog or class haar features.Pattern is known
Other method is SVM or neutral net, adaboost.Input picture detects vehicle using vehicle identification mode first, then uses
Edge prior validation of information vehicle is present, and finally confirms vehicle symmetry.This patent application provides a kind of effectively high vehicle and known
Not rate and the method for low vehicle false alarm rate.The method takes computing resource and time-consuming big, is not suitable for being transplanted to DSP.
It is entitled in publication number CN 101814238A《Loop Vehicle Detector and vehicle checking method》's
A kind of Loop Vehicle Detector is proposed in application for a patent for invention.Outer feeder line circle is embedded to underground, and vehicle is in induction coil
During detecting domains, loop coil inductance value is changed, and vehicle in detecting domains is indicated by corresponding predetermined frequency of oscillation change
In the presence of or pass through；Vehicle is put down by carrying out vehicle count during coil, and by vehicle by the Time Calculation vehicle of front and rear coil
Equal speed.This method needs to destroy road surface, and cost is high.
The content of the invention
The technical problem to be solved in the present invention is that the drawbacks described above for prior art can be realized quickly there is provided one kind
Effective detection vehicle simultaneously counts the magnitude of traffic flow, the base that need not be destroyed original pavement structure, reduce secondary input and maintenance cost
In the detection method and device of the row array formula crossing array area vehicle flowrate of video mode.
The technical solution adopted for the present invention to solve the technical problems is：A kind of row array formula road based on video mode of construction
The detection method of mouth array area vehicle flowrate, draws virtual coil identifier, positioned at road on each track of row array formula crossing array area
Video detecting device is hung with monitoring bar on face, the video detecting device is located at the surface of road and angular field of view is covered
The virtual coil identifier is covered, methods described comprises the following steps：
A the background image of the virtual coil identifier) is extracted using video background update method, and as mould
Plate；
B the present image of the virtual coil identifier position) is obtained, and by itself and the virtual coil identifier
Background image carry out template matches；
C) judge whether the matching value that the template matches are obtained is more than first threshold, in this way, judge current state as nothing
Vehicle covers the virtual coil identifier, does not cover that count is incremented, performs step D)；Otherwise, it is determined that current state is to have vehicle
The virtual coil identifier is covered, count is incremented for covering, and counting is not covered and is set to 0, return to step A)；
D do not cover whether counting is more than the first setting frame number or whether the matching value is more than Second Threshold described in) judging,
In this way, step E is performed)；Otherwise, return to step A)；
E) judge that the covering is counted and whether be more than the second setting frame number, in this way, vehicle count increase performs step F)；
Otherwise, step F is performed)；
F) covering counting is set to 0, return to step A).
It is described virtual in the detection method of the row array formula crossing array area vehicle flowrate of the present invention based on video mode
Coil identifier is made using reflectorized material.
It is described virtual in the detection method of the row array formula crossing array area vehicle flowrate of the present invention based on video mode
Coil identifier is combination of the numerical digit 1,2,3,4 with being oriented to arrow.
In the detection method of the row array formula crossing array area vehicle flowrate of the present invention based on video mode, covered from vehicle
Cover the virtual coil identifier and complete a vehicle count to the process that the virtual coil identifier is covered without vehicle.
In the detection method of the row array formula crossing array area vehicle flowrate of the present invention based on video mode, using mixing
Gaussian Background modeling pattern generates the background image of the virtual coil identifier.
The invention further relates to a kind of detection method for realizing the above-mentioned row array formula crossing array area vehicle flowrate based on video mode
Device, on each track of row array formula crossing array area draw virtual coil identifier, in the monitoring bar on road surface hang
Video detecting device is hung with, the video detecting device is located at the surface of road and angular field of view covers the virtual coil mark
Know symbol, described device includes：
Background image extraction unit：Background for extracting the virtual coil identifier using video background update method
Image, and as template；
Matching unit：Present image for obtaining the virtual coil identifier position, and by itself and the void
The background image for intending coil identifier carries out template matches；
Matching value judging unit：For judging whether the matching value that the template matches are obtained is more than first threshold, in this way,
Judging current state, count is incremented without the vehicle covering virtual coil identifier, not cover；Otherwise, it is determined that current state is
There is vehicle to cover the virtual coil identifier, count is incremented for covering, and counting is not covered and is set to 0, return；
Counting judging unit is not covered：For judging that whether described do not cover is counted more than the first setting frame number or described
Whether it is more than Second Threshold with value；
Covering counts judging unit：Whether it is more than the second setting frame number for judging that the covering is counted；
Zero setting unit：For covering counting to be set to 0, return.
In device of the present invention, the virtual coil identifier is made using reflectorized material.
In device of the present invention, the virtual coil identifier is numerical digit 1,2,3,4 with being oriented to arrow
Combination.
In device of the present invention, cover described virtual from the vehicle covering virtual coil identifier to without vehicle
The process of coil identifier completes a vehicle count.
In device of the present invention, the virtual coil identifier is generated using mixed Gaussian background modeling mode
Background image.
Implement the present invention the row array formula crossing array area vehicle flowrate based on video mode detection method and device, with
Lower beneficial effect：Due to drawing virtual coil identifier on road surface, that is, identifier is drawn as virtual coil, by by void
The present image and background image for intending coil identifier carry out template matches and judge coverage mode, and detection vehicle whether there is, so
Post analysis vehicle enters virtual coil identifier to virtual coil identifier of going out, and to carry out vehicle count, realizes quick detection
Vehicle simultaneously counts the magnitude of traffic flow, therefore it can realize quick effective detection vehicle and count the magnitude of traffic flow, need not destroy original road
Face structure, the secondary input of reduction and maintenance cost.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
There is the accompanying drawing used required in technology description to be briefly described, it should be apparent that, drawings in the following description are only this
Some embodiments of invention, for those of ordinary skill in the art, on the premise of not paying creative work, can be with
Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is one implementation of detection method and device of the row array formula crossing array area vehicle flowrate of the invention based on video mode
The position view of video detecting device and virtual coil identifier in example；
Fig. 2 is the angular field of view and the relation schematic diagram of virtual coil identifier of video detecting device in the embodiment；
Fig. 3 is the schematic diagram of virtual coil identifier in the embodiment；
Fig. 4 is the flow chart of method in the embodiment；
Fig. 5 is the structural representation of device in the embodiment.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.It is based on
Embodiment in the present invention, it is every other that those of ordinary skill in the art are obtained under the premise of creative work is not made
Embodiment, belongs to the scope of protection of the invention.
, please in the detection method and device embodiment of row array formula crossing array area vehicle flowrate of the present invention based on video mode
Referring to Fig. 1, virtual coil identifier 11 is drawn on each track of row array formula crossing array area, in the monitoring bar on road surface
It is hung with video detecting device 10, that is to say, that video detecting device 10 is hung in the monitoring bar on road surface, and visual angle is faced
Virtual coil identifier 11.Fig. 1 is the position view of video detecting device and virtual coil identifier in the present embodiment, from figure
In 1 it can be seen that video detecting device 10 and the respective position of virtual coil identifier 11.It is noted that video detection is set
Standby 10 use video camera.
Fig. 2 is the angular field of view and the relation schematic diagram of virtual coil identifier of video detecting device in the present embodiment；Fig. 2
In, video detecting device 10 is located at the surface of road and angular field of view covering virtual coil identifier 11, that is to say, that video
The angular field of view of detection device 10 includes virtual coil identifier 11, for detecting that vehicle completes traffic statistics.In Fig. 2, dotted line
Round position is the angular field of view of video detecting device 10, and virtual coil identifier 11 must be at the visual angle of video detecting device 10
In the range of, video detecting device 10 is located at directly over road, and angular field of view can see two tracks, and in the absence of track it
Between vehicle mutually block.Video detecting device 10 can actually see how many tracks, be hidden so that vehicle is not present in angular field of view
Keep off premised on the virtual coil identifier in other tracks, can be 1 track or a plurality of track.
Fig. 3 is the schematic diagram of virtual coil identifier in the present embodiment, in Fig. 3, in the visual angle model of video detecting device 10
In enclosing, virtual coil identifier 11 is plotted on road surface using reflectorized material, virtual coil identifier 11 be numerical digit 1,2,
3rd, 4 combination with being oriented to arrow, refers to Fig. 2 and Fig. 3.
Fig. 4 is the flow chart of method in the present embodiment, in Fig. 4, the above-mentioned row array formula crossing array area car based on video mode
The detection method of flow comprises the following steps：
Step S01 extracts the background image of virtual coil identifier using video background update method, and as mould
Plate：In the present embodiment, the background image of stable virtual coil identifier is extracted using video background update method, and is made
For template, the template matches for the later stage.It is noted that set context update frame number in the present embodiment, when background more
When new frame number is more than the context update frame number of the setting, step S02 is performed；When the frame number of context update is less than or equal to the back of the body
When scape updates frame number, then proceed context update, next two field picture judge until being more than the background of above-mentioned setting more
New frame number.The above-mentioned context update frame number set is 100 frame.
Step S02 obtains the present image of virtual coil identifier position, and by itself and virtual coil identifier
Background image carries out template matches：In this step, the present image of virtual coil identifier position is obtained, and by itself and void
The background image for intending coil identifier carries out template matches.It is noted that in the present embodiment, preferably being made according to video sequence
The background image of virtual coil identifier is generated with mixed Gaussian background modeling mode, then passes through present image (current picture)
With the template matches of background image, judge currently to cover virtual coil with the presence or absence of vehicle.
Whether the matching value that the matching of step S03 judge templets is obtained is more than first threshold：Obtained according to template matches
With value, current covering state can be obtained.Specifically, in this step, whether the matching value that judge templet matching is obtained is more than the
One threshold value, if it is determined that result be yes, then perform step S05；Otherwise, step S04 is performed.In the present embodiment, by the first threshold
Value is labeled as L.
Step S04 judges current state to have vehicle to cover the virtual coil identifier, and count is incremented for covering, does not cover
Counting is set to 0：If above-mentioned steps S03 judged result is no, this step is performed.In this step, judge current state to have
Vehicle covers the virtual coil identifier, and count is incremented for covering, and counting is not covered and is set to 0, this step return to step S01 has been performed
Calculate next two field picture.
Step S05 judges current state, and without vehicle covering virtual coil identifier, not cover, count is incremented：If above-mentioned
Step S03 judged result is yes, then performs this step.In this step, judge that current state covers virtual coil as without vehicle
Identifier, does not cover that count is incremented, has performed this step, performs step S06.
Step S06 judges do not cover whether counting is more than the first setting frame number or whether matching value is more than Second Threshold：This
In step, judgement does not cover whether counting is more than the first setting frame number or whether matching value is more than Second Threshold, if it is determined that
As a result it is yes, then performs step S07；Otherwise, return to step S01 calculates next two field picture.In the present embodiment, frame is set by first
Number mark is that Second Threshold is labeled as into H.
Step S07 judges that covering is counted and whether is more than the second setting frame number：If above-mentioned steps S06 judged result is
It is then to perform this step.In this step, judge covering count whether be more than second setting frame number, if it is determined that result be it is yes,
Then perform step S08；Otherwise, step S09 is performed.The second setting frame number is labeled as m in the present embodiment.
The increase of step S08 vehicle counts：If above-mentioned steps S07 judged result is yes, i.e., covering, which is counted, is more than m, then
Perform this step.In this step, vehicle count increase.This step has been performed, step S09 is performed.It is noted that this implementation
In example, virtual coil identifier is covered to the process vehicle meter of completion that virtual coil identifier is covered without vehicle from vehicle
Number.Namely it is covered as a cycle according to video sequence, counts a vehicle from there is vehicle to cover no vehicle.
In the present embodiment, vehicle covering judges and vehicle count two parts algorithm logic isolated operation that as a result then intersecting should
With.When there is covering behavior, then cumulative statistics cover time (covering counting), although there may be accumulated time different
Situation, but judge whether that the logic covered is identical.When in the presence of not covering, the cover time (does not cover meter for accumulation
Number), if the cover time is not more than n frames or more multiframe persistently, then it is assumed that do not cover state effectively, otherwise need to increase template
The judgement of the matching value of matching, matching value is higher than Second Threshold, it is believed that do not cover effectively.It is effective effectively to not covering from covering
A cycle, counts a vehicle.
Step S09 is set to 0 counting is covered：If above-mentioned steps S07 judged result is no, that is, covers counting and be less than or wait
In m, then this step is performed.In this step, covering counting is set to 0.This step is performed, return to step S01 calculates next frame figure
Picture.
The method of the present invention can fast and effectively detect row array crossing array area vehicle and calculating vehicle flow.The present invention's
Front end is concentrated on to vehicle detection and statistics for method and speed is fast, reduces the delayed possibility of vehicle flow feedback of the information, can subtract
Work load is dredged at few traffic police scene, and effectively evades the unreasonable caused congestion risk of traffic light time.The method of the present invention
Based on unobstructed video detection vehicle, it can be expanded in the case of unobstructed according to effective coverage in the angular field of view of actual camera
Greatly 3 roads.Hardware coil is replaced by the virtual coil identifier on road surface, it is not necessary to destroy original pavement structure.The opposing party
The corresponding video in face can be provided by the existing platform interface of traffic police, reduce secondary input and maintenance cost.Therefore it can be realized soon
Fast effective detection vehicle simultaneously counts the magnitude of traffic flow, need not destroy original pavement structure, reduce secondary input and maintenance cost.
The present embodiment further relates to a kind of detection for realizing the row array formula crossing array area vehicle flowrate based on video mode as described above
The device of method, its structural representation is as shown in Figure 5.In the present embodiment, drawn on each track of row array formula crossing array area
Virtual coil identifier 11, video detecting device 10 is hung with the monitoring bar on road surface states video detecting device 10 and be located at
The surface of road and angular field of view covering virtual coil identifier 11.Virtual coil identifier 11 is made using reflectorized material,
Virtual coil identifier 11 is combination of the numerical digit 1,2,3,4 with being oriented to arrow.
In Fig. 5, the device includes background image extraction unit 21, matching unit 22, matching value judging unit 23, not covered
Count judging unit 24, covering and count judging unit 25 and zero setting unit 26；Wherein, background image extraction unit 21 is used to utilize
Video background update method extracts the background image of virtual coil identifier, and as template；Using mixed Gaussian background
Modeling pattern generates the background image of virtual coil identifier.Matching unit 22 is used to obtain virtual coil identifier position
Present image, and the background image of itself and virtual coil identifier is subjected to template matches；Matching value judging unit 23 is used for
Whether the matching value that judge templet matching is obtained is more than first threshold, in this way, judges that current state covers dummy line as without vehicle
Identifier is enclosed, does not cover that count is incremented；Otherwise, it is determined that current state is has vehicle to cover virtual coil identifier, covering, which is counted, to be added
1, counting is not covered and is set to 0, and is returned；Covering, which counts judging unit 24, to be used to judge not cover whether counting is more than the first setting frame number
Or whether matching value is more than Second Threshold；Covering, which counts judging unit 25, to be used to judge whether covering counting is more than the second setting frame
Number；Zero setting unit 26 is used to set to 0 covering counting, returns.In the present embodiment, virtual coil identifier is covered to without car from vehicle
Covering virtual coil identifier process complete a vehicle count.The device of the present invention can realize quick effective detection vehicle
And count the magnitude of traffic flow, original pavement structure need not be destroyed, secondary input and maintenance cost is reduced.
In a word, in the present embodiment, in the range of virtual coil identifier 11, vehicle present image does mould with background image
Plate is matched, and matching value thinks vehicle covering virtual coil identifier 11 less than or equal to first threshold, otherwise it is assumed that currently without car
Covering, by this method it is determined that whether the region of virtual coil identifier 11 is covered by vehicle.The present invention is by vehicle
Identification enumeration problem is converted to covering state-detection problem, by template matches, whether will cover State Transferring for template matches,
It is whether consistent with background image, vehicle count is effectively carried out by virtual coil identifier mode.
In vehicle flow of the statistics by the region of virtual coil identifier 11, the vehicle continuous cover time reaches necessarily
Value thinks that covering is effective.Using virtual coil identifier 11 as vehicle detection basis, virtual coil identifier 11 itself can be compared with
The good state for showing array area road, the abundant edge angle point of virtual coil identifier 11 etc. is characterized in that estimation vehicle occupies void
Intend the important indicator of the state of coil identifier 11.In Same Scene, vehicle can be preferably judged whether, and do not influence
The use of other marks of road surface.The present invention can realize quick effective detection vehicle and count the magnitude of traffic flow, need not destroy original road
Face structure, the secondary input of reduction and maintenance cost.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
God is with principle, and any modification, equivalent substitution and improvements made etc. should be included in the scope of the protection.
Claims (10)
1. a kind of detection method of the row array formula crossing array area vehicle flowrate based on video mode, it is characterised in that on row array formula road
Virtual coil identifier is drawn on each track of mouth array area, video detecting device is hung with the monitoring bar on road surface,
The video detecting device is located at the surface of road and angular field of view covers the virtual coil identifier, and methods described includes
Following steps：
A the background image of the virtual coil identifier) is extracted using video background update method, and as template；
B the present image of the virtual coil identifier position) is obtained, and by its back of the body with the virtual coil identifier
Scape image carries out template matches；
C) judge whether the matching value that the template matches are obtained is more than first threshold, in this way, judge current state as without vehicle
The virtual coil identifier is covered, does not cover that count is incremented, step D is performed)；Otherwise, it is determined that current state is to have vehicle covering
The virtual coil identifier, count is incremented for covering, and counting is not covered and is set to 0, return to step A)；
D do not cover whether counting is more than the first setting frame number or whether the matching value is more than Second Threshold described in) judging, such as
It is to perform step E)；Otherwise, return to step A)；
E) judge that the covering is counted and whether be more than the second setting frame number, in this way, vehicle count increase performs step F)；Otherwise,
Perform step F)；
F) covering counting is set to 0, return to step A).
2. the detection method of the row array formula crossing array area vehicle flowrate according to claim 1 based on video mode, its feature
It is, the virtual coil identifier is made using reflectorized material.
3. the detection method of the row array formula crossing array area vehicle flowrate according to claim 2 based on video mode, its feature
It is, the virtual coil identifier is combination of the numerical digit 1,2,3,4 with being oriented to arrow.
4. the detection method of the row array formula crossing array area vehicle flowrate according to claim 3 based on video mode, its feature
It is, cover the virtual coil identifier from vehicle completes once to the process that the virtual coil identifier is covered without vehicle
Vehicle count.
5. the detection of the row array formula crossing array area vehicle flowrate based on video mode according to Claims 1-4 any one
Method, it is characterised in that the background image of the virtual coil identifier is generated using mixed Gaussian background modeling mode.
6. a kind of detection method for realizing the row array formula crossing array area vehicle flowrate as claimed in claim 1 based on video mode
Device, it is characterised in that virtual coil identifier, the prison on road surface are drawn on each track of row array formula crossing array area
Video detecting device is hung with control lever, the video detecting device is located at the surface of road and angular field of view covers the void
Intend coil identifier, described device includes：
Background image extraction unit：Background for extracting the virtual coil identifier using video background update method
Picture, and as template；
Matching unit：Present image for obtaining the virtual coil identifier position, and by itself and the dummy line
The background image for enclosing identifier carries out template matches；
Matching value judging unit：For judging whether the matching value that the template matches are obtained is more than first threshold, in this way, judge
Count is incremented without the vehicle covering virtual coil identifier, not cover for current state；Otherwise, it is determined that current state is to have car
The covering virtual coil identifier, count is incremented for covering, and counting is not covered and is set to 0, and returns；
Counting judging unit is not covered：For judging that whether described do not cover is counted more than the first setting frame number or the matching value
Whether Second Threshold is more than；
Covering counts judging unit：Whether it is more than the second setting frame number for judging that the covering is counted；
Zero setting unit：For covering counting to be set to 0, return.
7. according to claim 6 realize the row array formula crossing array area car as claimed in claim 1 based on video mode
The device of the detection method of flow, it is characterised in that the virtual coil identifier is made using reflectorized material.
8. according to claim 7 realize the row array formula crossing array area car as claimed in claim 1 based on video mode
The device of the detection method of flow, it is characterised in that the virtual coil identifier is numerical digit 1,2,3,4 with being oriented to arrow
Combination.
9. according to claim 8 realize the row array formula crossing array area car as claimed in claim 1 based on video mode
The device of the detection method of flow, it is characterised in that cover the virtual coil identifier to without described in vehicle covering from vehicle
The process of virtual coil identifier completes a vehicle count.
10. the realization row array as claimed in claim 1 based on video mode according to claim 6 to 9 any one
The device of the detection method of formula crossing array area vehicle flowrate, it is characterised in that generated using mixed Gaussian background modeling mode described
The background image of virtual coil identifier.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710313072.5A CN106991812A (en) | 2017-05-05 | 2017-05-05 | The detection method and device of row array formula crossing array area vehicle flowrate based on video mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710313072.5A CN106991812A (en) | 2017-05-05 | 2017-05-05 | The detection method and device of row array formula crossing array area vehicle flowrate based on video mode |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106991812A true CN106991812A (en) | 2017-07-28 |
Family
ID=59417581
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710313072.5A Withdrawn CN106991812A (en) | 2017-05-05 | 2017-05-05 | The detection method and device of row array formula crossing array area vehicle flowrate based on video mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106991812A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108257391A (en) * | 2018-01-17 | 2018-07-06 | 成都上甲光电科技有限公司 | Current indicator based on bridge |
| CN108986465A (en) * | 2018-07-27 | 2018-12-11 | 深圳大学 | A kind of method of vehicle Flow Detection, system and terminal device |
| CN109658688A (en) * | 2017-10-11 | 2019-04-19 | 深圳市哈工大交通电子技术有限公司 | The detection method and device of access connection traffic flow based on deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3842952B2 (en) * | 2000-04-27 | 2006-11-08 | 三菱重工業株式会社 | Traffic flow measuring device |
| CN101510358A (en) * | 2009-03-20 | 2009-08-19 | 吉林大学 | Method and apparatus for processing real time statistical vehicle flowrate using video image |
| CN103177586A (en) * | 2013-03-05 | 2013-06-26 | 天津工业大学 | Machine-vision-based urban intersection multilane traffic flow detection method |
| CN103366581A (en) * | 2013-06-28 | 2013-10-23 | 南京云创存储科技有限公司 | Traffic flow counting device and counting method |
| CN106355903A (en) * | 2016-09-13 | 2017-01-25 | 枣庄学院 | Method for detecting vehicle flow of multiple lanes on basis of video analysis |
-
2017
- 2017-05-05 CN CN201710313072.5A patent/CN106991812A/en not_active Withdrawn
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3842952B2 (en) * | 2000-04-27 | 2006-11-08 | 三菱重工業株式会社 | Traffic flow measuring device |
| CN101510358A (en) * | 2009-03-20 | 2009-08-19 | 吉林大学 | Method and apparatus for processing real time statistical vehicle flowrate using video image |
| CN103177586A (en) * | 2013-03-05 | 2013-06-26 | 天津工业大学 | Machine-vision-based urban intersection multilane traffic flow detection method |
| CN103366581A (en) * | 2013-06-28 | 2013-10-23 | 南京云创存储科技有限公司 | Traffic flow counting device and counting method |
| CN106355903A (en) * | 2016-09-13 | 2017-01-25 | 枣庄学院 | Method for detecting vehicle flow of multiple lanes on basis of video analysis |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109658688A (en) * | 2017-10-11 | 2019-04-19 | 深圳市哈工大交通电子技术有限公司 | The detection method and device of access connection traffic flow based on deep learning |
| CN108257391A (en) * | 2018-01-17 | 2018-07-06 | 成都上甲光电科技有限公司 | Current indicator based on bridge |
| CN108986465A (en) * | 2018-07-27 | 2018-12-11 | 深圳大学 | A kind of method of vehicle Flow Detection, system and terminal device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105303883B (en) | A kind of Roadside Parking management system and method | |
| CN102254429B (en) | Video identification-based detection method of detection device of violation vehicles | |
| CN103914688B (en) | A kind of urban road differentiating obstacle | |
| CN105046985B (en) | The traffic control system of major trunk roads system-wide section based on big data | |
| CN103065470B (en) | Detection device for behaviors of running red light of vehicle based on machine vision with single eye and multiple detection faces | |
| CN104537841B (en) | Unlicensed vehicle violation detection method and detection system thereof | |
| CN109003338A (en) | A kind of Roadside Parking self-clocking charging method and device | |
| CN107016861A (en) | Traffic lights intelligent control system based on deep learning and intelligent road-lamp | |
| CN105096615B (en) | Signalling-unit-based adaptive optimization control system | |
| CN105575125B (en) | A kind of wagon flow video detecting analysis system | |
| CN104805784B (en) | Gate fare evasion detection system and gate fare evasion detection method | |
| CN203346814U (en) | Dual-lane integrated community intelligent lane barrier management system based on vehicle license plate recognition technology | |
| CN107543580A (en) | A kind of urban waterlogging monitoring early-warning system | |
| CN106991812A (en) | The detection method and device of row array formula crossing array area vehicle flowrate based on video mode | |
| CN106846813A (en) | The method for building urban road vehicle image data base | |
| CN106408956A (en) | Tunnel traffic congestion quick evacuation method and control system | |
| CN106781460B (en) | A kind of road section traffic volume state determines method and device | |
| CN107591018A (en) | A kind of open section parking management method and system | |
| CN103268703A (en) | Snapshot system for bus lane | |
| CN106601023A (en) | Indoor parking lot vehicle guiding and positioning system and method | |
| CN111882690A (en) | ETC multi-sensing information fusion track reduction high-speed charging method and system | |
| CN106469507A (en) | Dynamic green wave system and method automatically adjusted based on Real-Time Traffic Volume data | |
| CN107590999A (en) | A kind of traffic state judging method based on bayonet socket data | |
| CN109377757A (en) | The vehicle driving track extraction method of license plate identification data based on the rough error containing multi-source | |
| CN111627217A (en) | Method, device and equipment for judging vehicle illegal turning around and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170728 |
|
| WW01 | Invention patent application withdrawn after publication |