Summary of the invention
This application provides the setting methods and system of a kind of car flow information monitoring device, it is intended at least to a certain extent
Solve one of above-mentioned technical problem in the prior art.
To solve the above-mentioned problems, this application provides following technical solutions:
A kind of setting method of car flow information monitoring device, comprising:
Step a: determining initial area-of-interest, and calculates the proximity relations between each initial area-of-interest;
Step b: each initial area-of-interest is gradually merged according to the proximity relations, obtains new sense
Interest unit;
Step c: the quantity of monitoring device needed for information of vehicles monitors between the calculating new unit interested, and according to
Calculated result carries out the setting of the monitoring device.
The technical solution that the embodiment of the present application is taken further include: in the step a, the calculating is each initial interested
Proximity relations between region includes: the relationship of the spatial position feature and road network according to each initial area-of-interest, is established
The connection relationship of initial area-of-interest and given road network;And combine the spatial position feature calculation of initial area-of-interest any
Proximity relations between two or more initial area-of-interests.
The technical solution that the embodiment of the present application is taken further include: according to proximity relations to described each first in the step b
The interesting region that begins to flourish gradually is merged, comprising: according to the proximity relations between each initial area-of-interest, according to from the near to the distant
Principle, using hierarchy clustering method gradually by apart from least two or more than two closest initial area-of-interests into
Row merges.
The technical solution that the embodiment of the present application is taken further include: according to proximity relations to described each first in the step b
The interesting region that begins to flourish gradually is merged, further includes: is recalculated between new unit interested and other initial area-of-interests
Proximity relations, and will be new interested according to the proximity relations between new unit interested and other initial area-of-interests
Unit is merged with apart from least two or more than two closest initial area-of-interests.
The technical solution that the embodiment of the present application is taken further include: described to calculate new unit interested in the step c
Between information of vehicles monitor needed for monitoring device quantity further include: whether the quantity of monitoring device needed for judging is more than limitation item
Part re-execute the steps b if the quantity of required monitoring device is more than restrictive condition;Wherein, the restrictive condition is practical
The quantity of monitoring device.
A kind of another technical solution that the embodiment of the present application is taken are as follows: setting system of car flow information monitoring device, comprising:
Target determination module: for determining initial area-of-interest;
First computing module: for calculating the proximity relations between each initial area-of-interest;
Region merging technique module: for gradually being closed according to the proximity relations to each initial area-of-interest
And obtain new unit interested;
Second computing module: required monitoring device is monitored for calculating information of vehicles between the new unit interested
Quantity, and carry out according to calculated result the setting of the monitoring device.
The technical solution that the embodiment of the present application is taken further include: first computing module calculates each initial region of interest
Proximity relations between domain specifically: according to the relationship of the spatial position feature of each initial area-of-interest and road network, establish
The connection relationship of initial area-of-interest and given road network;And combine the spatial position feature calculation of initial area-of-interest any
Proximity relations between two or more initial area-of-interests.
The technical solution that the embodiment of the present application is taken further include: the region merging technique module is according to proximity relations to described each
A initial area-of-interest is gradually merged specifically: according to the proximity relations between each initial area-of-interest, according to
Principle from the near to the distant gradually will initially feel emerging apart from closest at least two or more than two using hierarchy clustering method
Interesting region merges.
The technical solution that the embodiment of the present application is taken further include: the region merging technique module is according to proximity relations to described each
A initial area-of-interest is gradually merged further include: recalculates new unit interested and other initial area-of-interests
Between proximity relations, and according to the proximity relations between new unit interested and other initial area-of-interests by new sense
Interest unit is merged with apart from least two or more than two closest initial area-of-interests.
The technical solution that the embodiment of the present application is taken further includes result judgment module, and the result judgment module is for judging
Whether the quantity of required monitoring device is more than restrictive condition, if the quantity of required monitoring device is more than restrictive condition, is passed through
Region merging technique module continues region merging technique;Wherein, the restrictive condition is the quantity of actual monitoring equipment.
Compared with the existing technology, the beneficial effect that the embodiment of the present application generates is: the car flow information of the embodiment of the present application
The setting method and system of monitoring device pass through the fine degree for reducing area-of-interest, to reduce the demand of monitoring device,
So that it is maximum to the ability of the vehicle OD information covering between area-of-interest, required number of devices is avoided beyond current budget
The case where.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only to explain the application, not
For limiting the application.
Analysis towards wagon flow OD information is intended to analyze vehicle between the area-of-interest that can attract vehicle parks
Transinformation.In view of the mobile dependence traffic network of vehicle, therefore, with the associated section of area-of-interest (i.e. by interested
Set out or into the section of area-of-interest in region) determine the quantity of monitoring device, it may be assumed that and area-of-interest is rough,
The quantity for being associated with section is fewer, and monitoring device quantity required for covering to vehicle OD information between area-of-interest is fewer.Base
In above-mentioned, the purpose of the application is directed to the setting method and system of a kind of car flow information monitoring device, can set in monitoring
In the case where standby limited amount, by calculating the proximity relations between each area-of-interest, accordingly to each area-of-interest
Gradually merged, so that the fine degree of area-of-interest is reduced, to reduce the demand of monitoring device, so as to interested
The ability of vehicle OD information covering between region is maximum, for based on vehicle OD information transport need analysis and management branch is provided
It holds.In the examples below, only by monitoring device be camera for be specifically described, it will be understood that the application other
In embodiment, monitoring device may be other photographing elements, detecting element etc., and be equally applicable in addition to vehicle OD information with
Outer other information of vehicles monitoring, for example, car speed monitoring etc..
Specifically, referring to Fig. 1, being the flow chart of the setting method of the car flow information monitoring device of the embodiment of the present application.
The setting method of the car flow information monitoring device of the embodiment of the present application the following steps are included:
Step 100: initial area-of-interest being determined according to actual purpose, and is calculated between each initial area-of-interest
Proximity relations;
In step 100, the mode for determining initial area-of-interest includes: according to hotel, scenic spot, cell and village and other places
Species type and road name, place name etc. determine initial area-of-interest.Calculate the proximity relations between each initial area-of-interest
Calculation are as follows: according to the relationship of the spatial position feature of each initial area-of-interest and road network, establish initial interested
The connection relationship in region and given road network;And according to actual needs, in conjunction with the spatial position feature calculation of initial area-of-interest
Proximity relations between arbitrary initial area-of-interest.
Among the above, in the embodiment of the present application, it is measured using the shortest distance of road network between initial area-of-interest each
Proximity relations between a initial area-of-interest.After initial area-of-interest and road network establish a connection, Ke Yigen
According to the shortest distance (i.e. the distance of shortest path) between the topological relationship calculation arbitrary node of road network connection.Due to needing to calculate
The distance between any two points, the embodiment of the present application (solve any two section in network structure using Floyd-Warshall algorithm
A kind of algorithm of shortest path between point) calculate the shortest distance.The algorithm basic thought is (node i and j) between two nodes
The shortest distance exist only in two kinds may: (1) directly from node i to j;(2) i arrives j by several nodes k again.If sharp
With Dis (i, j) indicate the distance between node i and j, to the node of all processes judge Dis (i, k)+Dis (k, j) < Dis (i,
J), if set up, Dis (i, j)=Dis (i, k)+Dis (k, j), after having traversed the node k of all processes, Dis are set
(i, j) is the shortest distance between two nodes.It is appreciated that minimum distance calculation side used in the embodiment of the present application
Method is one of proximity relations balancing method, in practical applications, can be extended according to actual needs and again fixed
Justice.
Step 200: according to the proximity relations between each initial area-of-interest according to certain principle to related initial sense
Interest region is gradually merged, and a certain number of units interested are formed;
In step 200, the combination principle that the embodiment of the present application gradually merges related initial area-of-interest are as follows:
According to the proximity relations between each initial area-of-interest, according to principle from the near to the distant, gradually using hierarchy clustering method
It will merge, formed a certain number of interested apart from least two or more than two closest initial area-of-interests
Unit;And then pay close attention to the wagon flow OD information between each unit interested.Wherein, single cluster side is generated with other cluster modes
Case is different, and what hierarchical clustering method generated is a cluster level, carries out not to initial area-of-interest in the embodiment of the present application
With the merging of degree.In view of combination principle from the near to the distant, the embodiment of the present application is using clustering method from bottom to top, constantly
Ground merges closest initial area-of-interest, forms new unit interested, until reaching setting target.
Further, referring to Fig. 2, being the side that the embodiment of the present application gradually merges related initial area-of-interest
The flow chart of method.Method that the embodiment of the present application gradually merges related initial area-of-interest the following steps are included:
Step 201: each initial area-of-interest is classified as one kind;
Step 202: according to the proximity relations between initial area-of-interest, by closest at least two or more than two
Initial area-of-interest merge, form new unit interested;
Step 203: recalculating the proximity relations between new unit interested and other initial area-of-interests, and root
It is according to the proximity relations between new unit interested and other initial area-of-interests that new unit interested and distance is most adjacent
At least two or more than two close initial area-of-interests merge, until meeting specific cluster condition.
In step 203, between the unit interested and other initial area-of-interests newly formed during hierarchical clustering
The calculating of proximity relations will have a direct impact on the result of cluster.According to the difference of distance calculating method, hierarchical clustering algorithm point
For SL method (single-linkage, single connection algorithm), CL method (complete-linkage, full-join algorithm) and AL method
(average-linkage, average level algorithm) three types, wherein class spacing is the spacing of classification inner element in SL method
From minimum value, and it is maximum value that the class spacing of CL method is corresponding, and the class spacing of AL method is then the spacing with classification inner element
From average level substitute, generally indicated with intermediate value.Due to AL method more can overall distance between reflecting regional, the application
Embodiment preferably uses AL method, but in the application other embodiments, is equally applicable to SL method and CL method.
Step 300: calculating the quantity of camera needed for vehicle OD information monitoring between each unit interested after merging;
In step 300, the calculation of required camera quantity are as follows: firstly, calculating and closing according to the connection relationship of road network
The association section of all units interested after and is (i.e. all to enter each unit interested and from each unit interested
The section set out), and required camera quantity is calculated according to the case where association section, pass through the prison in these association sections
Camera is arranged in point position, is monitored by camera to these association sections, vehicle between unit interested can be realized
The extraction of OD information.In the embodiment of the present application, setting single camera can only monitor on road from the side of camera direction
Carry out the license plate of all automobiles to traveling, and the road in survey region does not have the road of one-way trip, in such cases, often
The vehicle OD information monitoring in item association section at least needs 2 cameras.
Secondly, if the section of all directions of intersection belongs to the association section of unit interested, due to vehicle
Flow of the flow in road junction position is conservation, and the vehicle OD information in the one of section of intersection can lead to
The vehicle OD information for crossing other directions is estimated to obtain, and therefore, the camera in the section is redundancy, can remove.Please specifically it join
Fig. 3 is read, is that the intersection information of vehicles of the embodiment of the present application assesses schematic diagram.Wherein, IA and OA respectively indicates vehicle from A
Direction enters and goes out, and!IA and!OA respectively indicates vehicle and does not enter and go out from the direction A, other are analogized.So, if
The vehicles while passing situation in tri- directions known A, B and C, can calculate the vehicles passing in and out information in the direction D.Circular is such as
Under:
ID=(OA∪OB∪OC)∩(!IA∩!IB∩!IC) (1)
OD=(IA∪IB∪IC)∩(!OA∩!OB∩!OC) (2)
The vehicle pass-through situation that camera section is not configured in intersection can be calculated by formula (1) and (2), thus
Further reduce the quantity of camera.And the case where other types crossing, it can be calculated according to this principle.It can manage
It solves, the calculation method of camera quantity is without being limited thereto in the embodiment of the present application, can also specifically be extended according to actual needs.
Step 400: whether the quantity of camera needed for judging is more than restrictive condition, if the quantity of required camera is more than
Restrictive condition then re-execute the steps 200, until the quantity of required camera is no more than restrictive condition.
In step 400, according to the camera quantity and restrictive condition (the camera quantity of actual budget) being calculated
Compare, if meeting corresponding preset condition, stop union operation, on the association section of unit interested after merging
Camera is arranged in monitoring location, thus in the case where camera limited amount, so as to the vehicle between unit interested
OD information covering power maximizes.And support the setting location schemes of camera under different camera quantity restrictive conditions formulation and
Selection, can be used for comparing the fine degree of the vehicle OD information extraction under different schemes, and it is more scientific reasonable to be conducive to formulate
Practical application goal programming.
Referring to Fig. 4, being the structural schematic diagram of the setting system of the car flow information monitoring device of the embodiment of the present application.This Shen
Please the setting system of car flow information monitoring device of embodiment include target determination module, the first computing module, region merging technique mould
Block, the second computing module and result judgment module.
Target determination module: for determining initial area-of-interest according to actual purpose;Wherein it is determined that initial region of interest
The mode in domain includes: determining initial interested according to the types of ground objects such as hotel, scenic spot, cell and village and road name, place name etc.
Region.
First computing module: for calculating the proximity relations between each initial area-of-interest;Wherein, it calculates each first
The calculation of proximity relations between beginning area-of-interest are as follows: according to the spatial position feature of each initial area-of-interest with
The relationship of road network establishes the connection relationship of initial area-of-interest Yu given road network;And according to actual needs, in conjunction with initially feeling emerging
Proximity relations between the spatial position feature calculation arbitrary initial area-of-interest in interesting region.
Among the above, measured using the shortest distance of road network between initial area-of-interest each initial area-of-interest it
Between proximity relations.After initial area-of-interest and road network establish a connection, the topology that can be connected according to road network is closed
System calculates the shortest distance between arbitrary node.Due to needing to calculate the distance between any two points, the embodiment of the present application is used
Floyd-Warshall algorithm calculates the shortest distance.The algorithm basic thought is the most short distance of (node i and j) between two nodes
It is possible from existing only in two kinds: (1) directly from node i to j;(2) i arrives j by several nodes k again.If using Dis (i,
J) the distance between node i and j are indicated, Dis (i, k)+Dis (k, j) < Dis (i, j) is judged to the node of all processes, if
It sets up, then sets Dis (i, j)=Dis (i, k)+Dis (k, j), after having traversed the node k of all processes, Dis (i, j) is just
It is the shortest distance between two nodes.It is appreciated that minimum distance calculation method used in the embodiment of the present application is
One of proximity relations balancing method can be extended and redefine according to actual needs in practical applications.
Region merging technique module: for according to the proximity relations between each initial area-of-interest according to certain principle to phase
It closes initial area-of-interest gradually to be merged, forms a certain number of units interested;Wherein, the embodiment of the present application is to correlation
The combination principle that initial area-of-interest is gradually merged are as follows: according to the proximity relations between each initial area-of-interest,
It, gradually will be initial apart from closest at least two or more than two using hierarchy clustering method according to principle from the near to the distant
Area-of-interest merges, and forms a certain number of units interested;And then pay close attention to the wagon flow between each unit interested
OD information.Wherein, generate that single clustering schemes are different from other cluster modes, what hierarchical clustering method generated is a cluster layer
It is secondary, different degrees of merging is carried out to initial area-of-interest in the embodiment of the present application.In view of merging from the near to the distant is former
Then, the embodiment of the present application constantly merges closest initial area-of-interest using clustering method from bottom to top,
New unit interested is formed, until reaching setting target.
Further, the merging that region merging technique module gradually merges related initial area-of-interest specifically includes:
A: each initial area-of-interest is classified as one kind;
B: initial by closest at least two or more than two according to the proximity relations between initial area-of-interest
Area-of-interest merges, and forms new unit interested;
C: the proximity relations between new unit interested and other initial area-of-interests is recalculated, and according to new
Proximity relations between unit interested and other initial area-of-interests by new unit interested with apart from it is closest extremely
Few two or more initial area-of-interests merge, until meeting specific cluster condition;Wherein, level
The calculating of proximity relations between the unit interested newly formed in cluster process and other initial area-of-interests can direct shadow
Ring the result to cluster.According to the difference of distance calculating method, hierarchical clustering algorithm is divided into SL method (single-linkage, list
Join algorithm), CL method (complete-linkage, full-join algorithm) and AL method (average-linkage, average level
Algorithm) three types, wherein the minimum value of class spacing distance between classification inner element in SL method, and the class spacing of CL method
Corresponding is maximum value, and the class spacing of AL method is substituted with the average level of distance between classification inner element, general to use
Intermediate value indicates.Due to AL method more can overall distance between reflecting regional, the embodiment of the present application preferably uses AL method, but at this
Apply being equally applicable to SL method and CL method in other embodiments.
Second computing module: it is imaged needed for vehicle OD information monitoring between each unit interested after merging for calculating
The quantity of head;The calculation of camera quantity are as follows: firstly, the institute's thoughts calculated after merging are emerging according to the connection relationship of road network
The association section of interesting unit, and required camera quantity is calculated according to the case where association section, by these association roads
Camera is arranged in the monitoring location of section, is monitored by camera to these association sections, unit interested can be realized
Between vehicle OD information extraction.In the embodiment of the present application, setting single camera can only monitor on road from camera court
To direction running and carry out the license plates of all automobiles, and the road in survey region does not have the road of one-way trip, such feelings
Under condition, the vehicle OD information monitoring in every association section at least needs 2 cameras.
Secondly, if the section of all directions of intersection belongs to the association section of unit interested, due to vehicle
Flow of the flow in road junction position is conservation, and the vehicle OD information in the one of section of intersection can lead to
The vehicle OD information for crossing other directions is estimated to obtain, and therefore, the camera in the section is redundancy, can remove.Please specifically it join
Fig. 3 is read, is that the intersection information of vehicles of the embodiment of the present application assesses schematic diagram.Wherein, IA and OA respectively indicates vehicle from A
Direction enters and goes out, and!IA and!OA respectively indicates vehicle and does not enter and go out from the direction A, other are analogized.So, if
The vehicles while passing situation in tri- directions known A, B and C, can calculate the vehicles passing in and out information in the direction D.Circular is such as
Under:
ID=(OA∪OB∪OC)∩(!IA∩!IB∩!IC) (1)
OD=(IA∪IB∪IC)∩(!OA∩!OB∩!OC) (2)
The vehicle pass-through situation that camera section is not configured in intersection can be calculated by formula (1) and (2), thus
Further reduce the quantity of camera.And the case where other types crossing, it can be calculated according to this principle.It can manage
It solves, the calculation method of camera quantity is without being limited thereto in the embodiment of the present application, can also specifically be extended according to actual needs.
As a result judgment module: for judging whether the quantity of required camera is more than restrictive condition, if required camera
Quantity be more than restrictive condition, then again through region merging technique module carry out region merging technique, until required camera quantity not
Until restrictive condition;In device, according to the camera quantity and restrictive condition (the camera number of actual budget being calculated
Amount) compare, if meeting corresponding preset condition, stop union operation, the association section of unit interested after merging
On monitoring location camera is set, thus in the case where camera limited amount, so as between unit interested
Vehicle OD information covering power maximizes.And support the system of camera setting location schemes under different camera quantity restrictive conditions
Fixed and selection, can be used for comparing the fine degree of the vehicle OD information extraction under different schemes, is conducive to formulate more scientific
The goal programming of reasonable practical application.
In order to verify the feasibility of the application, following embodiment is carried out by research case of Dapeng peninsula of Shenzhen City region
Test.Dapeng peninsula possesses multiple natural scenic spots, while also including town dweller accumulation regions and villages within the city.According to Dapeng peninsula
The case where, it is initial region of interest that the embodiment of the present application, which selectes the types such as natural scenic spot, villages within the city and town dweller's cell,
Domain, and the purpose that camera is laid is then to monitor the wagon flow OD situation between each initial area-of-interest.According to this Shen
Please wagon flow OD information monitoring equipment in embodiment setting method, respectively in the feelings of given 10,30 and 50 cameras
Under condition, the final amalgamation result of corresponding initial area-of-interest and need to be arranged the section that camera is observed information of vehicles
As shown in figure 5, being result schematic diagram of the embodiment of the present application in Dapeng peninsula regional experiment.Above-mentioned experimental result intuitively provides
Under the restrictive condition of the camera of different number, the section of when camera setting towards wagon flow OD information required monitoring,
The crossing of respective stretch or on the way position can be used as the foundation of camera setting.
The setting method and system of the car flow information monitoring device of the embodiment of the present application reduce the fine journey of area-of-interest
Degree, to reduce the demand of monitoring device, so that it is maximum to the ability of the vehicle OD information covering between area-of-interest, it avoids
Required number of devices exceeds the case where current budget, overcomes and is currently based in road traffic features information monitoring equipment more
The problem of paying close attention to key road segment, slipping so as to cause vehicle OD information;Meanwhile the application can according to the camera of different number into
The formulation and optimization of the row plan of establishment are conducive to make for comparing the fine degree of the vehicle OD information extraction under different schemes
The goal programming of fixed more scientific reasonable practical application.
The foregoing description of the disclosed embodiments makes professional and technical personnel in the field can be realized or use the application.
Various modifications to these embodiments will be readily apparent to those skilled in the art, as defined herein
General Principle can be realized in other embodiments without departing from the spirit or scope of the application.Therefore, the application
It is not intended to be limited to the embodiments shown herein, and is to fit to and the principles and novel features disclosed herein phase one
The widest scope of cause.