CN107204114A - A kind of recognition methods of vehicle abnormality behavior and device - Google Patents

A kind of recognition methods of vehicle abnormality behavior and device Download PDF

Info

Publication number
CN107204114A
CN107204114A CN201610156687.7A CN201610156687A CN107204114A CN 107204114 A CN107204114 A CN 107204114A CN 201610156687 A CN201610156687 A CN 201610156687A CN 107204114 A CN107204114 A CN 107204114A
Authority
CN
China
Prior art keywords
control point
vehicle
point position
tested
driving trace
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Withdrawn
Application number
CN201610156687.7A
Other languages
Chinese (zh)
Inventor
马昌军
解海波
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
ZTE Corp
Original Assignee
ZTE Corp
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by ZTE Corp filed Critical ZTE Corp
Priority to CN201610156687.7A priority Critical patent/CN107204114A/en
Priority to PCT/CN2017/073477 priority patent/WO2017157119A1/en
Publication of CN107204114A publication Critical patent/CN107204114A/en
Withdrawn legal-status Critical Current

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
    • G08G1/0125—Traffic data processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
    • G06V30/10—Character recognition
    • G06V30/19—Recognition using electronic means
    • G06V30/196—Recognition using electronic means using sequential comparisons of the image signals with a plurality of references
    • G06V30/1983—Syntactic or structural pattern recognition, e.g. symbolic string recognition
    • G06V30/1988—Graph matching
    • 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
    • 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
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/08—Detecting or categorising vehicles

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Multimedia (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Traffic Control Systems (AREA)

Abstract

The invention provides a kind of recognition methods of vehicle abnormality behavior and device, recognition methods includes：The vehicle monitoring data of the multiple control points position collection set in statistical regions are obtained, and from the behavioural characteristic data of vehicle monitoring extracting data vehicle to be tested；According to behavioural characteristic data, the driving trace sequence of vehicle to be tested is obtained；According to the vehicle behavior pattern in driving trace sequence and the statistical regions obtained in advance, judge that vehicle to be tested whether there is abnormal behaviour.The present invention solves the defect that the vehicle monitoring data that can not be collected in the prior art to multiple control points are associated analysis, solve simultaneously the problem of can not be to vehicle abnormality behavior automatic decision in the long period, add recognition accuracy and the recognition efficiency of vehicle abnormality behavior.

Description

A kind of recognition methods of vehicle abnormality behavior and device
Technical field
The present invention relates to intelligent traffic monitoring field, more particularly, to a kind of recognition methods of vehicle abnormality behavior And device.
Background technology
The reality that the detection process and traffic conditions for carrying out break in traffic rules and regulations using traffic information collection technology are monitored in real time Border with relatively broad, prior art can be realized to making a dash across the red light, driving over the speed limit, parking offense and The illegal activities such as reverse driving carry out automatic detection and evidence obtaining of taking pictures.In addition, in the long period or Vehicle abnormality behavior in spatial extent is such as：Block number plate, conversion number plate and deck etc., in addition it is also necessary to many The time discontinuous data that individual camera is shot are associated analysis, for such demand, current common side Case has：
First, carrying out manual information retrieval and artificial filter：For example fake license plate vehicle is recognized, typically the true car of number plate Main discovery vehicle exists abnormal violating the regulations and excessive to vehicle supervision department's report or fake license plate vehicle number of times violating the regulations Traffic control department is caused to note, manual information retrieval goes out all monitoring records of the number plate, and the artificial examination hair of contrast one by one Existing vehicles peccancy.But when manually carrying out manual information retrieval and filtering, efficiency is very limited, it is impossible to taken the photograph to multiple As the magnanimity monitoring data of head is associated analysis.
Second, the time-space relationship by analyzing vehicle, i.e. vehicle are judged by minimum passing time between bayonet socket Whether deck：If a vehicle is less than some time threshold (correspondence in the time difference that two bayonet sockets are taken Minimum passage time between bayonet socket), then it is assumed that wherein one car is fake-licensed car.But, by between bayonet socket most Small transit time judges the mode of deck, and its is difficult to accurate setting minimum passage time, there is larger erroneous judgement Probability, can not detect if occurring in addition when fake license plate vehicle is different from real vehicles, there is certain fail to judge Probability.
Although both the above scheme can find some abnormal behaviours of vehicle, both the above scheme is deposited In following shortcoming, wherein, when artificial progress manual information retrieval and filtering, efficiency is very limited, it is impossible to multiple The magnanimity monitoring data of camera is associated analysis.In addition, being judged by minimum passing time between bayonet socket The mode of deck, its minimum passage time is difficult to accurate setting, there is larger probability of miscarriage of justice, if covered in addition Occur then detecting when board vehicle is different from real vehicles, there is certain probability of failing to judge.If in addition, Vehicle is blocked or changed number plate within certain a period of time to car plate, by charge station, traffic checking road Mouthful before number plate is gained into true car plate again, change board process unless photographed, otherwise the class behavior be difficult to find or Judge.
To sum up judge, prior art has the vehicle monitoring data correlation point that can not be photographed to multiple cameras The problem of analysis, i.e., judgement can not be identified to the vehicle abnormality behavior in the long period.
The content of the invention
In order to solve asking for judgement can not to be identified to the vehicle abnormality behavior in the long period in the prior art Topic, the invention provides a kind of recognition methods of vehicle abnormality behavior and device.
In order to solve the above-mentioned technical problem, the invention provides a kind of recognition methods of vehicle abnormality behavior, institute Stating recognition methods includes：
The vehicle monitoring data of the multiple control points position collection set in statistical regions are obtained, and from the vehicle The behavioural characteristic data of vehicle to be tested are extracted in monitoring data；
According to the behavioural characteristic data, the driving trace sequence of the vehicle to be tested is obtained；
According to the vehicle behavior pattern in the driving trace sequence and the statistical regions obtained in advance, sentence The vehicle to be tested that breaks whether there is abnormal behaviour.
Optionally, the behavioural characteristic data include the number-plate number and control point bit identification, described in the basis Behavioural characteristic data, obtain the driving trace sequence of the vehicle to be tested, including：According to the vehicle to be tested The number-plate number and the vehicle to be tested by the time sequencing of multiple control points positions in statistical regions, will be to be tested The corresponding control point bit identification in multiple control points position that vehicle is sequentially passed through is recorded as the traveling rail of the vehicle to be tested Mark sequence.
Optionally, the driving trace sequence according to the vehicle to be tested and the Statistical Area obtained in advance Vehicle behavior pattern in domain, judge the vehicle to be tested whether there is abnormal behaviour before, the identification side Method also includes：According to the vehicle monitoring data of all vehicles in the statistical regions gathered in advance, car is obtained Behavior pattern, wherein, the vehicle behavior pattern includes：Reachability relation, monitoring between control point position The driving trace class template of transit time and vehicle between point position.
Optionally, the reachability relation between the control point position is determined in such a way：According to the statistics The vehicle monitoring data of all vehicles in region, obtain the driving trace sequence of all vehicles；To described The driving trace sequence of all vehicles carries out statistical analysis, obtains normally travel track sets, and determine normal Adjacent control point bit identification in driving trace sequence；According to the adjacent control point bit identification, had The control point position of direct reachability relation, wherein, the previous control point that the adjacent control point bit identification is indicated Position can be directly to up to the latter control point position in adjacent control point bit identification.
Optionally, the transit time between the control point position is determined in such a way：According to the vehicle Monitoring data, calculates average current between the control point position with direct reachability relation in a preset time period During current between time, and the control point position that the average transit time is recorded as in the preset time period Between.
Optionally, it is described according to the behavioural characteristic data, obtain the driving trace sequence of the vehicle to be tested Afterwards, the recognition methods also includes：Number-plate number wrong data to the vehicle to be tested is modified, Specifically include：According to the reachability relation between control point position, the driving trace of the vehicle to be tested is judged Whether there is direct reachability relation between the corresponding control point position of adjacent control point bit identification in sequence；If not having Have, then obtain all paths between the control point position without direct reachability relation in the statistical regions, Wherein, the sequence that path is made up of the corresponding control point bit identification in control point position sequentially passed through is represented；According to Control point bit identification in all paths, obtains in all paths and monitors abnormal monitoring data Control point bit identification, wherein, the abnormal monitoring data at least include：The number-plate number and vehicle monitored Register information is not inconsistent, without the corresponding control point of adjacent control point bit identification in number-plate number register information, path Position is unsatisfactory for control point without the control point position in direct reachability relation and path with direct reachability relation Transit time between position；The control point bit identification for monitoring abnormal monitoring data is incorporated into and described waits to validate the car Driving trace sequence in, and judge adjacent control point bit identification correspondence in the driving trace sequence after merging Control point position whether meet control point position between transit time；If meet, and the license plate number monitored The similitude of code and the car plate number plate of the vehicle to be tested is more than the first preset value, then by the license plate number monitored Code is modified.
Optionally, the driving trace sequence according to the vehicle to be tested and the Statistical Area obtained in advance Vehicle behavior pattern in domain, judges that the vehicle to be tested whether there is abnormal behaviour, including：According to described The transit time between reachability relation and control point position between driving trace sequence, control point position, judges The vehicle to be tested whether there is abnormal behaviour；And/or according to the driving trace sequence and driving trace class mould Plate, judges that the vehicle to be tested whether there is abnormal behaviour.
Optionally, the reachability relation and monitoring according between the driving trace sequence, control point position Transit time between point position, judges that the vehicle to be tested whether there is abnormal behaviour, including：According to described The transit time between reachability relation and control point position between driving trace sequence, control point position, successively Judge whether the corresponding control point position of adjacent control point bit identification has in the driving trace sequence of the vehicle to be tested There is direct reachability relation and whether meet the transit time between control point position；If adjacent control point bit identification Corresponding control point position does not have direct reachability relation, or the corresponding control point position of adjacent control point bit identification With direct reachability relation but be unsatisfactory for control point position between transit time, then judge that the vehicle to be tested is deposited In abnormal behaviour.
Optionally, it is described according to the driving trace sequence and driving trace class template, judge described in wait to validate the car Whether there is abnormal behaviour, including：According to a prefixed time interval, by the traveling rail of the vehicle to be tested Mark sequence is split as multiple driving trace subsequences；According to the first control point of driving trace subsequence position Mark and last position control point bit identification, obtaining has the first control point bit identification and last position control point position mark The driving trace class template of knowledge；Calculate the driving trace subsequence and the driving trace class template obtained Between similitude, if the similitude is less than the second preset value, judge that the vehicle to be tested is present abnormal Behavior.
According to another aspect of the present invention, present invention also offers a kind of identifying device of vehicle abnormality behavior, The identifying device includes：
First acquisition module, the vehicle monitoring of the multiple control points position collection set for obtaining in statistical regions Data, and from the behavioural characteristic data of the vehicle monitoring extracting data vehicle to be tested；
Second acquisition module, for according to the behavioural characteristic data, obtaining the traveling rail of the vehicle to be tested Mark sequence；
Judge module, for according to the car in the driving trace sequence and the statistical regions obtained in advance Behavior pattern, judges that the vehicle to be tested whether there is abnormal behaviour.
Optionally, the behavioural characteristic data include the number-plate number and control point bit identification, and described second obtains Module is specifically for according to the number-plate number of the vehicle to be tested and the vehicle to be tested by statistical regions The time sequencing of multiple control point positions, the corresponding control point in multiple control points position that vehicle to be tested is sequentially passed through Bit identification is recorded as the driving trace sequence of the vehicle to be tested.
Optionally, the identifying device also includes the 3rd acquisition module, for according to the system gathered in advance The vehicle monitoring data of all vehicles in region are counted, vehicle behavior pattern is obtained, wherein, the vehicle behavior Pattern includes：Control point position between reachability relation, control point position between transit time and vehicle row Sail track class template.
Optionally, the 3rd acquisition module is specifically for according to the car of all vehicles in the statistical regions Monitoring data, obtains the driving trace sequence of all vehicles；To the driving trace of all vehicles Sequence carries out statistical analysis, obtains normally travel track sets, and determine the phase in normally travel track sets Adjacent control point bit identification；According to the adjacent control point bit identification, the monitoring with direct reachability relation is obtained Point position, wherein, the previous control point position that the adjacent control point bit identification is indicated can be directly to up to adjacent prison Latter control point position in control point bit identification.
Optionally, the 3rd acquisition module is also particularly useful for according to the vehicle monitoring data, calculating one Average transit time in preset time period between the control point position with direct reachability relation, and will be described flat Equal transit time is recorded as the transit time between the control point position in the preset time period.
Optionally, the identifying device also includes correcting module, for the number-plate number to the vehicle to be tested Wrong data is modified, specifically for according to the reachability relation between control point position, judging described Whether have between the corresponding control point position of adjacent control point bit identification in the driving trace sequence of vehicle to be tested straight Connect reachability relation；If not having, the control point for not having direct reachability relation in the statistical regions is obtained All paths between position, wherein, path is by the corresponding control point bit identification group in control point position that sequentially passes through Into sequence represent；According to the control point bit identification in all paths, obtain in all paths and supervise The control point bit identification of abnormal monitoring data is controlled, wherein, the abnormal monitoring data at least include：Monitoring To the number-plate number with vehicle registration information be not inconsistent, without adjacent control point in number-plate number register information, path The corresponding control point position of bit identification is without the prison in direct reachability relation and path with direct reachability relation Control point position is unsatisfactory for the transit time between control point position；The control point position mark of abnormal monitoring data will be monitored Knowledge is incorporated into the driving trace sequence of the vehicle to be tested, and judges phase in the driving trace sequence after merging Whether the corresponding control point position of adjacent control point bit identification meets the transit time between control point position；If meeting, And the number-plate number monitored and the similitude of the car plate number plate of the vehicle to be tested are more than the first preset value, Then the number-plate number monitored is modified.
Optionally, the judge module is specifically for according between the driving trace sequence, control point position Reachability relation and control point position between transit time, judge the vehicle to be tested with the presence or absence of exception row For；And/or according to the driving trace sequence and driving trace class template, judge whether the vehicle to be tested is deposited In abnormal behaviour.
Optionally, the judge module include the first judging unit, for according to the driving trace sequence, The transit time between reachability relation and control point position between control point position, waits to validate the car described in judgement successively Driving trace sequence in the corresponding control point position of adjacent control point bit identification whether there is direct reachability relation And whether meet control point position between transit time；If the corresponding control point position of adjacent control point bit identification Have directly up to pass without direct reachability relation, or the corresponding control point position of adjacent control point bit identification The transit time for being but being unsatisfactory between control point position, then judge that the vehicle to be tested has abnormal behaviour.
Optionally, the judge module also includes the second judging unit, for according to a prefixed time interval, The driving trace sequence of the vehicle to be tested is split as multiple driving trace subsequences；According to the traveling rail The first control point bit identification of mark subsequence and last position control point bit identification, obtaining has the first control point The driving trace class template of bit identification and last position control point bit identification；The driving trace subsequence is calculated with obtaining Similitude between the driving trace class template taken, if the similitude is less than the second preset value, sentences There is abnormal behaviour in the fixed vehicle to be tested.
The beneficial effects of the invention are as follows：
A kind of recognition methods for vehicle abnormality behavior that the present invention is provided, obtains what is set in statistical regions first The vehicle monitoring data of multiple control points position collection, and from the behavior of vehicle monitoring extracting data vehicle to be tested Characteristic, then according to behavioural characteristic data, obtains the driving trace sequence of vehicle to be tested, finally according to Driving trace sequence and the vehicle behavior pattern obtained in advance, judge that vehicle to be tested whether there is abnormal behaviour. The analysis for the driving trace sequence that the present invention is validated the car by treating, solving in the prior art can not be to multiple The vehicle monitoring data that control point is collected are associated the defect of analysis, at the same solve can not to it is longer when In vehicle abnormality behavior automatic decision the problem of, add the recognition accuracy and knowledge of vehicle abnormality behavior Other efficiency.
Brief description of the drawings
Fig. 1 represents the step flow chart of the recognition methods of vehicle abnormality behavior in the first embodiment of the present invention；
Fig. 2 represents to obtain the step flow of the reachability relation between control point position in the second embodiment of the present invention Figure；
Fig. 3 represents the control point position arrangement schematic diagram of a traffic intersection；
Fig. 4 represents the reachability relation figure between the control point of each in Fig. 3 position；
Fig. 5 represents to treat the number-plate number wrong data validated the car in the third embodiment of the present invention and is modified Step flow chart；
Fig. 6 represents the structured flowchart of the identifying device of vehicle abnormality behavior in the fifth embodiment of the present invention.
Embodiment
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although being shown in accompanying drawing The exemplary embodiment of the disclosure, it being understood, however, that may be realized in various forms the disclosure without should be by Embodiments set forth here is limited.It is opposite to be able to be best understood from this there is provided these embodiments It is open, and can by the scope of the present disclosure completely convey to those skilled in the art.
First embodiment：
As shown in figure 1, to be flowed in the first embodiment of the present invention the step of recognition methods of vehicle abnormality behavior Cheng Tu, the recognition methods includes：
Step 101, the vehicle monitoring data of the multiple control points position collection set in acquisition statistical regions, and from The behavioural characteristic data of vehicle monitoring extracting data vehicle to be tested.
In this step, the multiple control points position set in statistical regions can shoot the image by vehicle, And the image of vehicle is handled and optical identification, the vehicle monitoring data of collection vehicle.Specifically, prison Control point position uninterrupted sampling passes through the vehicle image of control point position, and extracts the number-plate number of vehicle, vehicle The characteristic informations such as model, vehicle color, the monitoring data storehouse set up in statistical regions.Optionally, number is monitored It can not only include by the number-plate number of the vehicle of control point position and by supervising according to the vehicle monitoring data in storehouse The dynamic data such as passage time of control point position, in addition to the control point bit identification of control point position, control point position The static datas such as position.In addition, in this step, can also be from vehicle monitoring extracting data vehicle to be tested Behavioural characteristic data, wherein, behavioural characteristic data at least include：The number-plate number and control point position are corresponding Control point bit identification.
Step 102, according to behavioural characteristic data, the driving trace sequence of vehicle to be tested is obtained.
Specifically, behavioural characteristic data include the number-plate number and control point bit identification, in this step, can be with Pass through the time sequencing of multiple control point positions in statistical regions according to the number-plate number of vehicle to be tested and vehicle to be tested, The corresponding control point bit identification in multiple control points position that vehicle to be tested is sequentially passed through is recorded as the vehicle to be tested Driving trace sequence.
Step 103, according to the vehicle behavior pattern in driving trace sequence and the statistical regions obtained in advance, sentence The vehicle to be tested that breaks whether there is abnormal behaviour.
In this step, the vehicle behavior pattern in statistical regions can be obtained in advance, and then basis is waited to validate the car Driving trace sequence and the statistical regions that obtain in advance in vehicle behavior pattern, judge that vehicle to be tested is It is no to there is abnormal behaviour.Specifically, the vehicle behavior pattern in statistical regions can include supervising in statistical regions Control point position between reachability relation, control point position between transit time and vehicle driving trace class template. The abnormal behaviour of vehicle can include the row such as vehicle fake-license, vehicle board turning, shielding automobile number plate and illegal running For.
The driving trace sequence that the present embodiment is validated the car by treating is analyzed, and according to driving trace sequence Vehicle behavior pattern in row and the statistical regions obtained in advance, judges that vehicle to be tested whether there is abnormal behaviour, The vehicle monitoring data that solving can not collect to multiple control points in the prior art are associated lacking for analysis Fall into, while solving the problem of can not be to vehicle abnormality behavior automatic decision in the long period, add and treat The recognition accuracy for an abnormal behaviour of validating the car and recognition efficiency.
Second embodiment：
Vehicle behavior pattern according to driving trace sequence and the statistical regions obtained in advance, judges to be tested Vehicle whether there is before abnormal behaviour, in addition it is also necessary to obtain the vehicle behavior pattern in statistical regions.Specifically, Vehicle behavior pattern can be obtained according to the vehicle monitoring data of all vehicles in the statistical regions gathered in advance, Wherein, vehicle behavior pattern includes：Control point position between reachability relation, control point position between it is current when Between and vehicle driving trace class template.In addition, when collecting vehicle monitoring data in real time, can be by The vehicle monitoring data obtained in real time are equally as the foundation for obtaining vehicle behavior pattern.
Vehicle behavior pattern is explained below.
Specifically, vehicle can be shot by multiple control points position on the way in the process of moving, vehicle monitoring is formed Data.By the analysis and research of the history vehicle monitoring data to magnanimity it can be found that the motion of vehicle has one Fixed time and space idea, these rules then form certain vehicle behavior pattern.Specifically, can will be each Control point position regards a state as, then the corresponding control point bit identification in all control points position is constituted in statistical regions One finite state space, the finite state space can be expressed as E={ 1,2 ..., N }.When vehicle is different Control point position when photographing, being considered as vehicle is shifted between different states, wherein, one Secondary vehicle travel process can be described using the corresponding control point bit identification in control point position of vehicle process, example Such as, A vehicles, which once travel the driving trace sequence of process, can use A=(A1,A2,…,An) be indicated.So Afterwards according to Statistics, it can be deduced that transition probability of the vehicle between different conditions.So, according to advance The vehicle monitoring data of all vehicles in the statistical regions of collection, can obtain the reachable pass between control point position The vehicle behavior patterns such as the driving trace class template of transit time and vehicle between system, control point position.Tool Body, when obtaining vehicle behavior pattern, it can also sample and choose the vehicle prison of Some vehicles in statistical regions Data are controlled, but in order to avoid the inaccuracy of sampling, the vehicle prison of all vehicles in preferred pair statistical regions Control data are analyzed, and draw vehicle behavior pattern.
In turn below to logical between the reachability relation between control point position in vehicle behavior pattern, control point position The determination method of the driving trace class template of row time and vehicle is illustrated.
As shown in Fig. 2 the step to determine the reachability relation between control point position in the second embodiment of the present invention Rapid flow chart, including：
Step 201, according to the vehicle monitoring data of all vehicles in statistical regions, the traveling of all vehicles is obtained Track sets.
In this step, specifically, can be obtained according to the vehicle monitoring data of all vehicles in statistical regions The behavioural characteristic data of all vehicles are taken, then according to the behavioural characteristic data of all vehicles, all cars are obtained Driving trace sequence, i.e., it is corresponding with control point that the vehicle is sequentially passed through position according to the number-plate number of vehicle Control point bit identification, obtain the driving trace sequence of the vehicle.
Step 202, the driving trace sequence to all vehicles carries out statistical analysis, obtains normally travel track sequence Row, and determine the adjacent control point bit identification in normally travel track sets.
In this step, specifically, when carrying out statistical analysis to the driving trace sequences of all vehicles, such as When really the occurrence number of a traveling track sets is less than a pre-determined threshold, then it is assumed that the driving trace sequence is different Normal driving trace sequence；Otherwise it is assumed that the driving trace sequence is normally travel track sets.Specifically, The pre-determined threshold can be determined according to the total quantity of driving trace sequence and total species of driving trace sequence. For example, can be obtained according to the ratio of the total quantity of driving trace sequence and total species of driving trace sequence The average occurrence number of one traveling track sets, then can set the pre- gating according to the average occurrence number Limit, such as, according to average occurrence number and the product of a proportionality coefficient, determine the pre-determined threshold.Optionally, The proportionality coefficient can be the numerical value less than 1, such as 0.1,0.01 etc..
Step 203, according to adjacent control point bit identification, the control point position with direct reachability relation is obtained.
In this step, the monitoring with direct reachability relation can be obtained according to adjacent control point bit identification Point position, specifically, the previous control point position that adjacent control point bit identification is indicated can be directly to up to adjacent monitoring Latter control point position in point bit identification.
In addition, making description below to the control point position with direct reachability relation：If the first control point position can The second control point position is directly reached not via the 3rd control point position, then can determine the first control point position and the Two control points position has direct reachability relation；If the first control point position must could be arrived via the 3rd control point position Up to the second control point position, it is determined that the first control point position and the second control point position do not have direct reachability relation. Wherein, the first control point position, the second control point position and the 3rd control point position are different, and are Statistical Area Any control point position in domain in multiple control point positions.
The reachability relation between control point position is illustrated below.
As shown in figure 3, being the control point position arrangement schematic diagram of a traffic intersection.In figure 3, A, B, C, D, E, F, G and H are the control point bit identification of control point position, and from figure 3, it can be seen that road 1 No turns.Assuming that a normally travel track sets are FGHC, then it can draw, adjacent control point bit identification F and G, G and H, H and the corresponding control point positions of C are the control point position with direct reachability relation.According to It is secondary to analogize, it can be deduced that the adjacent control point bit identification in all normally travel track sets, and draw each Reachability relation between control point position.Specifically, as shown in figure 4, between each control point position in Fig. 3 Reachability relation figure, wherein, the oriented line between two control point bit identifications represents two control points position mark Knowing between corresponding control point position has direct reachability relation, in addition, oriented same two monitoring of expression of line Transfer path between point position is oriented.
Specifically, after the reachability relation between getting control point position, matrix can also be utilized P=(pij)n×n, record the reachability relation between control point position.Wherein, P represent control point position between can Up to relational matrix, n represents the corresponding control point bit identification in all control points position, p in statistical regionsijRepresent P The corresponding control point position of i-th row control point bit identification control point corresponding with jth row control point bit identification in matrix Reachability relation between position.Wherein, p is worked asijWhen=1, the i-th row control point bit identification correspondence in P matrixes is represented Control point position corresponding with jth row control point bit identification control point position with direct reachability relation；Work as pij=0 When, represent the corresponding control point position of the i-th row control point bit identification and jth row control point bit identification pair in P matrixes The control point position answered does not have direct reachability relation.In addition, herein it should be noted that in a matrix, two Reachability relation between the corresponding control point position of individual control point bit identification is oriented, for example, in pijIn, i-th The previous control point position that bit identification corresponding control point position in row control point indicates for adjacent control point bit identification, The latter control point that bit identification corresponding control point position in jth row control point indicates for adjacent control point bit identification Position.
, optionally, can be with it is determined that during transit time between control point position in addition, in the present embodiment According to vehicle monitoring data, calculate in a preset time period between the control point position with direct reachability relation It is logical between average transit time, and the control point position that average transit time is recorded as in the preset time period The row time.
Specifically, because the time that vehicle passes through between the control point position with direct reachability relation is with traffic Changed condition, i.e., the transit time between control point position has Tide Characteristics, therefore can be according to different pre- If the period, the average transit time between the control point position with direct reachability relation is calculated respectively.For example, It was that can remain with certain lap between a preset time period, and different preset time periods with 70 minutes, Then can be by 0：00~1：10 are divided into a preset time period, 1：00~2：10 be a preset time Section, the like, it can be deduced that different preset time periods.
Certainly, after the transit time between getting control point position, record can equally be used a matrix to The transit time between control point position in one preset time period.For example, it is possible to use matrix T=(tij)n×n, Record the transit time between the control point position in a preset time period.Wherein, T is represented in a preset time period Control point position between transit time matrix, n represents the corresponding control point in all control points position in statistical regions Bit identification, tijRepresent the corresponding control point position of the i-th row control point bit identification and jth row control point position in T matrixes Identify the transit time between corresponding control point position.Wherein, when the i-th row control point bit identification pair in T matrixes When the control point position control point position corresponding with jth row control point bit identification answered has direct reachability relation, tij's It is worth for transit time；When the corresponding control point position of the i-th row control point bit identification in T matrixes and jth row control point When the corresponding control point position of bit identification does not have direct reachability relation, tijValue for infinity.
In addition, in the present embodiment, optionally, the driving trace of the vehicle in vehicle behavior pattern is obtained During class template, clustering algorithm can be used, clustering is carried out to all normally travel track sets, obtained The driving trace class template of vehicle.
Specifically, using clustering algorithm, clustering is carried out to all normally travel track sets, obtained During the driving trace class template of vehicle, Map in mapping-reduction (map-reduce) method can be used Function and Reduce functions obtain the driving trace sequence of vehicle：In Map functions, each control point is obtained Each vehicle monitoring data of position, and the number-plate number of a vehicle is passed through the vehicle as key values The other informations such as time, control point bit identification, vehicle model and the vehicle color of each control point position are used as value Value passes to Reduce functions；, can be to the data of same key values according to warp in Reduce functions The time of mistake is ranked up, so as to obtain the driving trace sequence of the vehicle.In addition, obtaining the traveling of vehicle During the class template of track, it is possible to use Hausdorff distance (hausdorff distance) measures all driving traces The distance between corresponding driving trace of sequence, then seeks the track similarity between driving trace.It is assumed that If m bar driving trace sequences, then a similarity matrix S=(S can be obtainedij)m×m, wherein, SijTable Show the track phase between the corresponding driving trace of the i-th row and the corresponding driving trace of jth row in similarity matrix S Like degree.Finally, after the track similarity of all driving trace sequences is obtained, it is possible to use spectral clustering is calculated The corresponding driving trace of m bar driving trace sequences is divided into q driving trace class template by method, it is preferred that poly- Class algorithm can use K averages (K-means) clustering algorithm.
The present embodiment is carried out according to the vehicle monitoring data of all vehicles in statistical regions to vehicle behavior pattern Training, adds the accuracy of vehicle behavior pattern, so that according to the driving trace sequence of vehicle to be tested With vehicle behavior pattern, when judging that vehicle to be tested whether there is abnormal behaviour, an abnormal behaviour of validating the car is treated Identification is more accurate.
3rd embodiment：
In the vehicle monitoring data of the control point position collection got, due to various factors such as picture quality Not high, image recognition mistake or network transmission etc., may cause vehicle monitoring data division mistake or not Completely, for example, " D " and " 0 " in the number-plate number, number relatively such as " L ", " T " and " 1 " Code has certain probability to will recognise that error result.Such partial error or incomplete vehicle monitoring data meeting Have influence on the identification of acquisition and the vehicle abnormality behavior of vehicle behavior pattern.Therefore, according to behavioural characteristic After data, the driving trace sequence for obtaining vehicle to be tested, in addition it is also necessary to treat the number-plate number mistake validated the car Data are modified.If specifically, adjacent control point bit identification pair in the driving trace sequence of vehicle to be tested Do not have direct reachability relation between the control point position answered, but disclosure satisfy that current between control point position Time, then the vehicle may by way of control point position there occurs the number-plate number recognize mistake, cause vehicle to be supervised Control or is incorporated into loss of data by error message, now by carrying out examination filtering to abnormal monitoring data, can be with Realization is modified to number-plate number wrong data.
Specifically, as shown in figure 5, wrong to treat the number-plate number validated the car in the third embodiment of the present invention The step flow chart that data are modified by mistake, including：
Step 301, according to the reachability relation between control point position, in the driving trace sequence for judging vehicle to be tested Whether adjacent control point bit identification has direct reachability relation between corresponding control point position.
In this step, it is assumed that the driving trace sequence of vehicle to be tested is A=(A1,A2,…,An), then can root According to the reachability relation between control point position, judge that adjacent control point bit identification is corresponding in driving trace sequence A Whether there is direct reachability relation between control point position, if without direct reachability relation, into step 302.
Step 302, if not having, the control point position without direct reachability relation in statistical regions is obtained Between all paths.
In this step, if specifically, adjacent control point bit identification pair in the driving trace sequence of vehicle to be tested Do not have direct reachability relation between the control point position answered, then without directly up to pass in acquisition statistical regions All paths between the control point position of system, wherein, path can be corresponding by the control point position sequentially passed through The sequence of control point bit identification composition is represented.Such as in driving trace sequence A adjacent control point bit identification A1 And A2Do not have direct reachability relation between corresponding control point position, then obtain adjacent control point bit identification A1 And A2All paths between corresponding control point position.Assuming that path has a m bars, and in the i-th paths from A1Reach A2Need altogether by z control point position, then path P={ P1,P2,……,Pm, path Pi=(A1,Ai1,……,Aiz,A2), wherein, m represents adjacent control point bit identification A1And A2Corresponding control point The bar number in all paths, P between positioniRepresent adjacent control point bit identification A1And A2Between corresponding control point position The i-th paths, AizRepresent adjacent control point bit identification A1And A2I-th between corresponding control point position Control point bit identification in paths.
Step 303, the control point bit identification in all paths, obtains in all paths and monitors abnormal prison Control the control point bit identification of data.
In this step, specifically, the number-plate number that abnormal monitoring data at least include monitoring is stepped on vehicle Note information is not inconsistent, without the corresponding control point position of adjacent control point bit identification in number-plate number register information, path Control point position is unsatisfactory for without the control point position with direct reachability relation in direct reachability relation and path Between transit time, i.e., ought exist control point position monitor the number-plate number with register vehicle information be not inconsistent, or The number-plate number register information that person does not monitor, or the corresponding monitoring of adjacent control point bit identification in path Without the corresponding control point position of adjacent control point bit identification in direct reachability relation, or path between point position Between have direct reachability relation but be unsatisfactory for control point position between transit time when, obtain those control points The corresponding control point bit identification in position.
Step 304, the control point bit identification for monitoring abnormal monitoring data is incorporated into the traveling rail of vehicle to be tested In mark sequence, and judge the corresponding control point position of adjacent control point bit identification in the driving trace sequence after merging Whether transit time control point position between is met.
In this step, if specifically, adjacent control point bit identification correspondence in driving trace sequence after merging Control point position meet control point position between transit time, then into step 305.
Step 305, if the corresponding control point position of adjacent control point bit identification is full in driving trace sequence after merging Transit time between sufficient control point position, and the number-plate number monitored and the car plate number plate of vehicle to be tested Similitude is more than the first preset value, then is modified the number-plate number monitored.
In this step, if the corresponding control point of adjacent control point bit identification in driving trace sequence after merging The transit time between control point position, and the number-plate number monitored and the car plate of vehicle to be tested are met between position The similitude of number plate is more than the first preset value, then is modified the number-plate number monitored.Optionally, exist After the number-plate number monitored is modified, it can submit and manually be confirmed.In addition, for easily leading Cause wrong data number-plate number character can the artificial similar matching character of configuration file so that according to identification Data statistics draws correct characters.
For example, " Soviet Union A23F45 " and " the 5th character difference, and " F " in Soviet Union's number-plate numbers of A23P45 " two Similar to " P " profile, when having partial occlusion or other reasonses, F or P may be identified mistake.If The corresponding control point position of adjacent control point bit identification has direct reachability relation in driving trace sequence after merging, And the transit time between control point position is met, then number-plate number wrong data can be modified and be carried Hand over manual confirmation or re-recognize the number-plate number after carrying out the processing such as denoising to image.
In the present embodiment, it is modified by treating the number-plate number wrong data validated the car, it is ensured that treat The accuracy for the driving trace sequence validated the car, so as to improve the accuracy of vehicle abnormality Activity recognition to be tested.
Fourth embodiment：
, can after the vehicle behavior pattern in the driving trace sequence and statistical regions of vehicle to be tested is got According to the vehicle behavior pattern in driving trace sequence and the statistical regions obtained in advance, to judge vehicle to be tested With the presence or absence of abnormal behaviour.Specifically, according to driving trace sequence and the statistical regions obtained in advance Vehicle behavior pattern, judge vehicle to be tested whether there is abnormal behaviour when, can according to driving trace sequence, The transit time between reachability relation and control point position between control point position, judges whether vehicle to be tested is deposited In abnormal behaviour, whether vehicle to be tested can also be judged according to driving trace sequence and driving trace class template There is abnormal behaviour.
The two ways to above-mentioned judgement vehicle to be tested with the presence or absence of abnormal behaviour is explained below.
First, according between the reachability relation between driving trace sequence, control point position and control point position Transit time, judge vehicle to be tested whether there is abnormal behaviour when, can first according to driving trace sequence, The transit time between reachability relation and control point position between control point position, judges vehicle to be tested successively In driving trace sequence the corresponding control point position of adjacent control point bit identification whether have direct reachability relation and Whether transit time control point position between is met.If the corresponding control point position of adjacent control point bit identification does not have Have a direct reachability relation, or the corresponding control point position of adjacent control point bit identification have direct reachability relation but The transit time between control point position is unsatisfactory for, then judges that vehicle to be tested has abnormal behaviour.Optionally, exist This can also set a confidential interval for the transit time between control point position, when adjacent control point bit identification pair When there is direct reachability relation but transit time to exceed the confidential interval for the control point position answered, vehicle to be tested is judged There is abnormal behaviour.
Below to according between the reachability relation between driving trace sequence, control point position and control point position Transit time, judges that vehicle to be tested is illustrated with the presence or absence of the principle of abnormal behaviour.
In control point position under normal circumstances, vehicle is travelled on road to be collected by multiple control points position, this The corresponding control point bit identification in control point position constitutes the driving trace sequence of vehicle a bit.Due under normal circumstances The space transfer of vehicle is continuous, i.e., vehicle will not suddenly disappear from somewhere, and suddenly appear in strange land and Without vehicle monitoring data on the way, therefore the adjacent control point bit identification pair in the driving trace sequence of vehicle Necessarily there is direct reachability relation between the control point position answered, if without direct reachability relation, in vehicle In the case that abnormal monitoring data are not present in monitoring data, it is possible to determine that the vehicle has abnormal behaviour.In addition, Directly may be used if having between the corresponding control point position of adjacent control point bit identification in the driving trace sequence of vehicle Up to relation, but vehicle is unsatisfactory for prison by the transit time of the corresponding control point position of adjacent control point bit identification Transit time between control point position, then equally can be determined that vehicle has abnormal behaviour.
According to the method, to vehicle fake-license, vehicle board turning and the progress of the abnormal behaviours such as the number-plate number can be blocked Identification.
Vehicle fake-license abnormal behaviour is identified：
For example, the vehicle monitoring data for treating an A that validates the car are ranked up, obtain waiting to validate the car in certain time period A driving trace sequence is P=(A1,A2,…,An), adjacent monitoring in driving trace sequence P is judged successively Whether there is direct reachability relation between the corresponding control point position of point bit identification and whether meet control point position Between transit time.If P(i,i+1)=0, i.e. control point bit identification AiAnd Ai+1Between corresponding control point position not With direct reachability relation, then it can be designated as once abnormal；If control point bit identification AiAnd Ai+1Corresponding prison Control point has direct reachability relation between position, but vehicle passes through control point bit identification AiAnd Ai+1Corresponding prison The time t of control point position(i,i+1)It is unsatisfactory for the transit time between control point position, such as time t(i,i+1)Positioned at monitoring Outside 95% confidential interval of the transit time between point position, then it can equally be designated as once abnormal.Last root Sorted according to abnormal data ratio, it can be deduced that all vehicles are sorted by the possibility of deck.
Assuming that the driving trace sequence of fake license plate vehicle is Pfake=(A1,A2,…,An), the traveling of true car plate vehicle Track sets are Preal=(B1,B2,…,Bn).If fake license plate vehicle and true car plate vehicle occur, to car simultaneously Monitoring data is according to being likely to be obtained following sequence P after time-sequencingmix=(A1,A2,B1,A3,B2,B3,…,Bn,An)。 If fake license plate vehicle and true car plate vehicle running section are close in such cases, and vehicle passes through adjacent monitoring The time of the corresponding control point position of point bit identification meets the transit time between control point position, then now passes through prison Control point position between transit time can not find vehicle fake-license behavior, if but according to control point position between can Up to relation, it can be found that adjacent control point bit identification A2And B1, B1And A3, A3And B2Corresponding prison Control point does not have direct reachability relation between position, so as to draw sequence PmixThere is abnormal row in corresponding vehicle For, and due to sequence PmixThe same number-plate number of correspondence, then draws PmixCorresponding vehicle there may be set Board abnormal behaviour.In addition, if fake license plate vehicle occurs when different with true car plate vehicle, then fake license plate vehicle and The driving trace sequence of true car plate vehicle is P after mergingmix=(A1,A2,…,An,B1,B2,…,Bn), if vehicle By control point bit identification AnAnd B1Between time meet control point position between transit time, then now lead to Whether the transit time None- identified vehicle crossed between control point position has deck abnormal behaviour, but is due to deck Vehicle is different with true car plate movable vehicle scope, if now control point bit identification AnAnd B1Corresponding monitoring Point position does not have direct reachability relation, then still may determine that PmixIt is abnormal that corresponding vehicle there may be deck Behavior.
To vehicle board turning and block number-plate number abnormal behaviour and be identified：
The vehicle monitoring data for treating an A that validates the car are ranked up, and obtain vehicle A to be tested in certain time period Driving trace sequence is P=(A1,A2,…,An), adjacent control point position mark in driving trace sequence P is judged successively Whether know has direct reachability relation between corresponding control point position and whether meets logical between control point position The row time.If it was found that control point bit identification A1And A2Without directly up to pass between corresponding control point position System, and vehicle A to be tested is in control point bit identification A1And A2Passage time difference at corresponding control point position For t1And t2, can now take t1To t2It is all by control point bit identification A in period1And A2It is corresponding Number-plate number L=(the L of control point position1,L2,……,Ln), and respectively to each number-plate number in t1Previous moment is extremely t2Vehicle monitoring data between later moment in time are analyzed.If vehicle LiIn t1Previous moment is to t2It is latter In driving trace sequence between moment, the corresponding control point position of adjacent control point bit identification is without directly may be used Up to relation, then by vehicle LiIt is classified as suspected vehicles.Certainly, if vehicle LiNumber-plate number None- identified, then Equally by vehicle LiIt is classified as suspected vehicles.Finally, all vehicle monitoring data further to suspected vehicles are entered Row characteristics of image is recognized, if there is outside vehicle feature such as vehicle, color etc. and vehicle A to be tested outside spy Levy and be consistent, then it is assumed that vehicle A to be tested has board turning or blocks the abnormal behaviours such as number plate, can now extract Vehicle A to be tested vehicle monitoring data carry out manual examination and verification.
Second, according to driving trace sequence and driving trace class template, judging vehicle to be tested with the presence or absence of different During Chang Hangwei, the driving trace sequence of vehicle to be tested can be split as multiple according to a prefixed time interval Driving trace subsequence；Then according to the first control point bit identification of driving trace subsequence and last position control point Bit identification, obtains the driving trace class template with the first control point bit identification and last position control point bit identification； The similitude between driving trace subsequence and the driving trace class template of acquisition is finally calculated, if similitude is small In the second preset value, then judge that vehicle to be tested has abnormal behaviour.
Below to according to driving trace sequence and driving trace class template, judging vehicle to be tested with the presence or absence of abnormal Behavior is explained.
Assuming that it is S={ S that m bar driving trace class templates are had in a statistical regions1,……,Sm, treat and validate the car A vehicle monitoring data are ranked up, and obtain the driving trace sequence of vehicle A to be tested in certain time period For P=(A1,A2,…,An).Now according to a prefixed time interval, such as 1 hour, by vehicle A's to be tested Driving trace sequence P=(A1,A2,…,An) it is split as multiple driving trace subsequences, such as P1=(A1,A2,…,Ai), P2=(Ai+1,Ai+2,…,Aj) ... ..., Pn=(Aj+1,Aj+2,…,An)；Then respectively according to the traveling rail split out Mark subsequence PiThe first control point bit identification and last position control point bit identification, finding has identical the first monitoring The driving trace class template S of point bit identification and last position control point bit identificationi；Driving trace is finally calculated respectively Sequence PiWith driving trace class template SiBetween similitude, can be with if similitude is less than the second preset value Judge that vehicle A to be tested has abnormal behaviour.The method such as using typical Hausdorff distance judged, Driving trace subsequence PiWith driving trace class template SiBetween similitude it is smaller, vehicle to be tested exists abnormal The possibility of behavior is bigger.
The present embodiment is sentenced according to the vehicle behavior pattern in driving trace sequence and the statistical regions obtained in advance The vehicle to be tested that breaks whether there is abnormal behaviour, solve multiple control points can not be collected in the prior art Vehicle monitoring data are associated the defect of analysis, can not be to the vehicle abnormality in the long period while solving The problem of behavior automatic decision, add recognition accuracy and the recognition efficiency of vehicle abnormality behavior to be tested.
5th embodiment：
As shown in fig. 6, being the structural frames of the identifying device of vehicle abnormality behavior in the fifth embodiment of the present invention Figure, the identifying device includes：
First acquisition module 401, the vehicle prison of the multiple control points position collection set for obtaining in statistical regions Data are controlled, and from the behavioural characteristic data of vehicle monitoring extracting data vehicle to be tested；
Second acquisition module 402, for according to behavioural characteristic data, obtaining the driving trace sequence of vehicle to be tested；
Judge module 403, for according to the vehicle behavior in driving trace sequence and the statistical regions obtained in advance Pattern, judges that vehicle to be tested whether there is abnormal behaviour.
Optionally, behavioural characteristic data include the number-plate number and control point bit identification, and the second acquisition module is specific For, according to the number-plate number of vehicle to be tested and vehicle to be tested by statistical regions multiple control points positions when Between order, the corresponding control point bit identification in multiple control points position that vehicle to be tested is sequentially passed through is recorded as this and treats The driving trace sequence validated the car.
Optionally, identifying device also includes the 3rd acquisition module, for according in the statistical regions gathered in advance The vehicle monitoring data of all vehicles, obtain vehicle behavior pattern, wherein, vehicle behavior pattern includes：Prison Control point position between reachability relation, control point position between transit time and vehicle driving trace class template.
Optionally, the 3rd acquisition module is specifically for according to the vehicle monitoring number of all vehicles in statistical regions According to the driving trace sequence of all vehicles of acquisition；Driving trace sequence to all vehicles carries out statistical analysis, Normally travel track sets are obtained, and determine the adjacent control point bit identification in normally travel track sets；Root According to adjacent control point bit identification, the control point position with direct reachability relation is obtained, wherein, adjacent control point The previous control point position that bit identification is indicated can be directly to up to latter control point in adjacent control point bit identification Position.
Optionally, the 3rd acquisition module is also particularly useful for according to vehicle monitoring data, one preset time of calculating Average transit time in section between the control point position with direct reachability relation, and will average transit time note Record as the transit time between the control point position in the preset time period.
Optionally, identifying device also includes correcting module, the number-plate number wrong data validated the car for treating It is modified, specifically for according to the reachability relation between control point position, judging the traveling rail of vehicle to be tested Whether there is direct reachability relation between the corresponding control point position of adjacent control point bit identification in mark sequence；If no Have, then obtain all paths between the control point position without direct reachability relation in statistical regions, its In, the sequence that path is made up of the corresponding control point bit identification in control point position sequentially passed through is represented；According to institute There is the control point bit identification in path, obtain the control point position mark that abnormal monitoring data are monitored in all paths Know, wherein, abnormal monitoring data at least include：The number-plate number that monitors is not inconsistent with vehicle registration information, Do not have without the corresponding control point position of adjacent control point bit identification in number-plate number register information, path and directly may be used Up in relation and path with direct reachability relation control point position be unsatisfactory for control point position between it is current when Between；The control point bit identification for monitoring abnormal monitoring data is incorporated into the driving trace sequence of vehicle to be tested, And judge whether the corresponding control point position of adjacent control point bit identification meets prison in the driving trace sequence after merging Transit time between control point position；If meet, and the number-plate number monitored and the license plate number of vehicle to be tested The similitude of board is more than the first preset value, then is modified the number-plate number monitored.
Optionally, judge module is specifically for according to the reachable pass between driving trace sequence, control point position Transit time between system and control point position, judges that vehicle to be tested whether there is abnormal behaviour；And/or according to Driving trace sequence and driving trace class template, judge that vehicle to be tested whether there is abnormal behaviour.
Optionally, judge module includes the first judging unit, for according to driving trace sequence, control point position Between reachability relation and control point position between transit time, the driving trace of vehicle to be tested is judged successively Whether whether the corresponding control point position of adjacent control point bit identification have direct reachability relation and meet in sequence Transit time between control point position；If the corresponding control point position of adjacent control point bit identification is without directly may be used There is direct reachability relation but prison is unsatisfactory for up to relation, or the corresponding control point position of adjacent control point bit identification Transit time between control point position, then judge that vehicle to be tested has abnormal behaviour.
Optionally, judge module also includes the second judging unit, for according to a prefixed time interval, will treat The driving trace sequence validated the car is split as multiple driving trace subsequences；According to the head of driving trace subsequence Position control point bit identification and last position control point bit identification, obtaining has the first control point bit identification and last position monitoring The driving trace class template of point bit identification；Calculate driving trace subsequence with obtain driving trace class template it Between similitude, if similitude be less than the second preset value, judge that vehicle to be tested has abnormal behaviour.
The above is the preferred embodiment of the present invention, it should be pointed out that come for the ordinary person of the art Say, some improvements and modifications can also be made under the premise of the principle of the present invention is not departed from, these improve and moistened Decorations are also within the scope of the present invention.

Claims (18)

1. a kind of recognition methods of vehicle abnormality behavior, it is characterised in that the recognition methods includes：
The vehicle monitoring data of the multiple control points position collection set in statistical regions are obtained, and from the vehicle The behavioural characteristic data of vehicle to be tested are extracted in monitoring data；
According to the behavioural characteristic data, the driving trace sequence of the vehicle to be tested is obtained；
According to the vehicle behavior pattern in the driving trace sequence and the statistical regions obtained in advance, sentence The vehicle to be tested that breaks whether there is abnormal behaviour.
2. recognition methods according to claim 1, it is characterised in that the behavioural characteristic data include The number-plate number and control point bit identification, it is described according to the behavioural characteristic data, obtain the vehicle to be tested Driving trace sequence, including：
According to the number-plate number of the vehicle to be tested and the vehicle to be tested by multiple control points in statistical regions The time sequencing of position, the corresponding control point bit identification in multiple control points position that vehicle to be tested is sequentially passed through is recorded For the driving trace sequence of the vehicle to be tested.
3. recognition methods according to claim 2, it is characterised in that described according to the vehicle to be tested Driving trace sequence and the statistical regions that obtain in advance in vehicle behavior pattern, judge described to be tested Vehicle whether there is before abnormal behaviour, and the recognition methods also includes：
According to the vehicle monitoring data of all vehicles in the statistical regions gathered in advance, vehicle behavior is obtained Pattern, wherein, the vehicle behavior pattern includes：Reachability relation, control point position between control point position Between transit time and vehicle driving trace class template.
4. recognition methods according to claim 3, it is characterised in that between the control point position can Determined in such a way up to relation：
According to the vehicle monitoring data of all vehicles in the statistical regions, the traveling of all vehicles is obtained Track sets；
Driving trace sequence to all vehicles carries out statistical analysis, obtains normally travel track sets, And determine the adjacent control point bit identification in normally travel track sets；
According to the adjacent control point bit identification, the control point position with direct reachability relation is obtained, wherein, The previous control point position that the adjacent control point bit identification is indicated can be directly to up in adjacent control point bit identification Latter control point position.
5. recognition methods according to claim 4, it is characterised in that logical between the control point position The row time determines in such a way：
According to the vehicle monitoring data, the control point in a preset time period with direct reachability relation is calculated Average transit time between position, and the averagely transit time is recorded as the monitoring in the preset time period Transit time between point position.
6. recognition methods according to claim 4, it is characterised in that described according to the behavioural characteristic After data, the driving trace sequence for obtaining the vehicle to be tested, the recognition methods also includes：
Number-plate number wrong data to the vehicle to be tested is modified, and is specifically included：
According to the reachability relation between control point position, in the driving trace sequence for judging the vehicle to be tested Whether adjacent control point bit identification has direct reachability relation between corresponding control point position；
If not having, obtain in the statistical regions between the control point position without direct reachability relation All paths, wherein, the sequence that path is made up of the corresponding control point bit identification in control point position sequentially passed through Represent；
According to the control point bit identification in all paths, obtain in all paths and monitor abnormal prison The control point bit identification of data is controlled, wherein, the abnormal monitoring data at least include：The license plate number monitored Code with register vehicle information be not inconsistent, it is corresponding without adjacent control point bit identification in number-plate number register information, path Control point position it is discontented without the control point position in direct reachability relation and path with direct reachability relation Transit time between sufficient control point position；
The control point bit identification for monitoring abnormal monitoring data is incorporated into the driving trace sequence of the vehicle to be tested In row, and judge in the driving trace sequence after merging whether is the corresponding control point position of adjacent control point bit identification Meet the transit time between control point position；
If meet, and the number-plate number monitored and the similitude of the car plate number plate of the vehicle to be tested are more than The number-plate number monitored, then be modified by the first preset value.
7. recognition methods according to claim 4, it is characterised in that described according to the vehicle to be tested Driving trace sequence and the statistical regions that obtain in advance in vehicle behavior pattern, judge described to be tested Vehicle whether there is abnormal behaviour, including：
According to logical between the reachability relation between the driving trace sequence, control point position and control point position The row time, judge that the vehicle to be tested whether there is abnormal behaviour；And/or
According to the driving trace sequence and driving trace class template, judge the vehicle to be tested with the presence or absence of different Chang Hangwei.
8. recognition methods according to claim 7, it is characterised in that described according to the driving trace The transit time between reachability relation and control point position between sequence, control point position, judges described to be tested Vehicle whether there is abnormal behaviour, including：
According to logical between the reachability relation between the driving trace sequence, control point position and control point position The row time, the corresponding prison of adjacent control point bit identification in the driving trace sequence of the vehicle to be tested is judged successively Whether control point position has direct reachability relation and whether meets the transit time between control point position；
If the corresponding control point position of adjacent control point bit identification does not have direct reachability relation, or adjacent monitoring The transit time that the corresponding control point position of point bit identification has direct reachability relation but is unsatisfactory between control point position, Then judge that the vehicle to be tested has abnormal behaviour.
9. recognition methods according to claim 7, it is characterised in that described according to the driving trace Sequence and driving trace class template, judge that the vehicle to be tested whether there is abnormal behaviour, including：
According to a prefixed time interval, the driving trace sequence of the vehicle to be tested is split as multiple traveling rails Mark subsequence；
According to the first control point bit identification of the driving trace subsequence and last position control point bit identification, obtain Driving trace class template with the first control point bit identification and last position control point bit identification；
The similitude between the driving trace subsequence and the driving trace class template of acquisition is calculated, if The similitude is less than the second preset value, then judges that the vehicle to be tested has abnormal behaviour.
10. a kind of identifying device of vehicle abnormality behavior, it is characterised in that the identifying device includes：
First acquisition module, the vehicle monitoring of the multiple control points position collection set for obtaining in statistical regions Data, and from the behavioural characteristic data of the vehicle monitoring extracting data vehicle to be tested；
Second acquisition module, for according to the behavioural characteristic data, obtaining the traveling rail of the vehicle to be tested Mark sequence；
Judge module, for according to the car in the driving trace sequence and the statistical regions obtained in advance Behavior pattern, judges that the vehicle to be tested whether there is abnormal behaviour.
11. identifying device according to claim 10, it is characterised in that the behavioural characteristic packet The number-plate number and control point bit identification are included, second acquisition module is specifically for according to the vehicle to be tested The number-plate number and the vehicle to be tested by the time sequencing of multiple control points positions in statistical regions, will be to be tested The corresponding control point bit identification in multiple control points position that vehicle is sequentially passed through is recorded as the traveling rail of the vehicle to be tested Mark sequence.
12. identifying device according to claim 11, it is characterised in that the identifying device also includes 3rd acquisition module, for the vehicle monitoring data according to all vehicles in the statistical regions gathered in advance, Vehicle behavior pattern is obtained, wherein, the vehicle behavior pattern includes：Control point position between reachability relation, The driving trace class template of transit time and vehicle between control point position.
13. identifying device according to claim 12, it is characterised in that the 3rd acquisition module tool Body is used for, and according to the vehicle monitoring data of all vehicles in the statistical regions, obtains all vehicles Driving trace sequence；Driving trace sequence to all vehicles carries out statistical analysis, obtains normally travel Track sets, and determine the adjacent control point bit identification in normally travel track sets；According to the adjacent prison Control point bit identification, obtains the control point position with direct reachability relation, wherein, the adjacent control point position mark Know the previous control point position indicated to can be directly to up to the latter control point position in adjacent control point bit identification.
14. identifying device according to claim 13, it is characterised in that the 3rd acquisition module is also Specifically for according to the vehicle monitoring data, calculating has direct reachability relation in a preset time period Average transit time between control point position, and the average transit time is recorded as in the preset time period Control point position between transit time.
15. identifying device according to claim 13, it is characterised in that the identifying device also includes Correcting module, is modified for the number-plate number wrong data to the vehicle to be tested, specifically for root According to the reachability relation between control point position, adjacent prison in the driving trace sequence of the vehicle to be tested is judged Whether control point bit identification has direct reachability relation between corresponding control point position；If not having, institute is obtained State all paths between the control point position without direct reachability relation in statistical regions, wherein, path by The sequence of the corresponding control point bit identification composition in control point position sequentially passed through is represented；According to all paths In control point bit identification, obtain the control point bit identification that abnormal monitoring data are monitored in all paths, Wherein, the abnormal monitoring data at least include：The number-plate number that monitors is not inconsistent with vehicle registration information, Do not have without the corresponding control point position of adjacent control point bit identification in number-plate number register information, path and directly may be used Up in relation and path with direct reachability relation control point position be unsatisfactory for control point position between it is current when Between；The control point bit identification for monitoring abnormal monitoring data is incorporated into the driving trace sequence of the vehicle to be tested In row, and judge in the driving trace sequence after merging whether is the corresponding control point position of adjacent control point bit identification Meet the transit time between control point position；If meet, and the number-plate number monitored is waited to validate the car with described Car plate number plate similitude be more than the first preset value, then the number-plate number monitored is modified.
16. identifying device according to claim 13, it is characterised in that the judge module is specifically used According to logical between the reachability relation between the driving trace sequence, control point position and control point position The row time, judge that the vehicle to be tested whether there is abnormal behaviour；And/or according to the driving trace sequence and Driving trace class template, judges that the vehicle to be tested whether there is abnormal behaviour.
17. identifying device according to claim 16, it is characterised in that the judge module includes the One judging unit, for according to the reachability relation between the driving trace sequence, control point position and monitoring Transit time between point position, judges adjacent control point position in the driving trace sequence of the vehicle to be tested successively Identify whether corresponding control point position has direct reachability relation and whether meet current between control point position Time；If the corresponding control point position of adjacent control point bit identification does not have direct reachability relation, or adjacent prison When control point bit identification corresponding control point position has direct reachability relation but is unsatisfactory for current between control point position Between, then judge that the vehicle to be tested has abnormal behaviour.
18. identifying device according to claim 16, it is characterised in that the judge module also includes Second judging unit, for according to a prefixed time interval, the driving trace sequence of the vehicle to be tested to be torn open It is divided into multiple driving trace subsequences；According to the first control point bit identification of the driving trace subsequence and end Position control point bit identification, obtains the traveling with the first control point bit identification and last position control point bit identification Track class template；Calculate the phase between the driving trace subsequence and the driving trace class template of acquisition Like property, if the similitude is less than the second preset value, judge that the vehicle to be tested has abnormal behaviour.
CN201610156687.7A 2016-03-18 2016-03-18 A kind of recognition methods of vehicle abnormality behavior and device Withdrawn CN107204114A (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201610156687.7A CN107204114A (en) 2016-03-18 2016-03-18 A kind of recognition methods of vehicle abnormality behavior and device
PCT/CN2017/073477 WO2017157119A1 (en) 2016-03-18 2017-02-14 Method and device for identifying abnormal behavior of vehicle

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610156687.7A CN107204114A (en) 2016-03-18 2016-03-18 A kind of recognition methods of vehicle abnormality behavior and device

Publications (1)

Publication Number Publication Date
CN107204114A true CN107204114A (en) 2017-09-26

Family

ID=59851442

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610156687.7A Withdrawn CN107204114A (en) 2016-03-18 2016-03-18 A kind of recognition methods of vehicle abnormality behavior and device

Country Status (2)

Country Link
CN (1) CN107204114A (en)
WO (1) WO2017157119A1 (en)

Cited By (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107622667A (en) * 2017-09-28 2018-01-23 广州烽火众智数字技术有限公司 A kind of adulterium number-plate number detection method and system
CN108022427A (en) * 2017-10-30 2018-05-11 深圳市赛亿科技开发有限公司 A kind of recognition methods of fake license plate vehicle and system
CN108280435A (en) * 2018-01-25 2018-07-13 盛视科技股份有限公司 A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation
CN108986474A (en) * 2018-08-01 2018-12-11 平安科技（深圳）有限公司 Fix duty method, apparatus, computer equipment and the computer storage medium of traffic accident
CN108986478A (en) * 2018-09-17 2018-12-11 公安部交通管理科学研究所 A method of fixed point screens illegal vehicle
CN109637122A (en) * 2018-10-19 2019-04-16 天津易华录信息技术有限公司 A kind of data processing method, equipment and system
CN109657621A (en) * 2018-12-21 2019-04-19 杨文鑫 Big data trajectory analysis platform
CN109767615A (en) * 2018-10-19 2019-05-17 江苏智通交通科技有限公司 Road network traffic flow key flow direction and critical path analysis method
CN109831744A (en) * 2017-11-23 2019-05-31 腾讯科技（深圳）有限公司 It is a kind of exception track recognizing method, device and storage equipment
CN109859492A (en) * 2019-03-27 2019-06-07 成都市公安科学技术研究所 A kind of system and method identifying fake-licensed car
CN110728842A (en) * 2019-10-23 2020-01-24 江苏智通交通科技有限公司 Abnormal driving early warning method based on reasonable driving range of vehicles at intersection
CN110807694A (en) * 2019-11-06 2020-02-18 广东泰骅科技信息服务有限公司 Talent sharing system
CN110874355A (en) * 2019-11-28 2020-03-10 北京以萨技术股份有限公司 Method, system, terminal and medium for detecting abnormal behavior of vehicle loitering winding
CN110942640A (en) * 2019-12-04 2020-03-31 无锡华通智能交通技术开发有限公司 Method for actively discovering suspect vehicle illegally engaged in network car booking passenger transportation
CN111179603A (en) * 2018-11-09 2020-05-19 杭州海康威视数字技术股份有限公司 Vehicle identification method and device, electronic equipment and storage medium
CN111191949A (en) * 2020-01-03 2020-05-22 北京三快在线科技有限公司 Method and device for identifying abnormal driving behaviors of networked vehicles and electronic equipment
CN111291216A (en) * 2020-02-28 2020-06-16 罗普特科技集团股份有限公司 Method and system for analyzing foothold based on face structured data
CN111367906A (en) * 2019-07-23 2020-07-03 杭州海康威视系统技术有限公司 Abnormal vehicle identification method, device, equipment and computer readable storage medium
CN111369805A (en) * 2020-01-09 2020-07-03 杭州海康威视系统技术有限公司 Fake plate detection method and device, electronic equipment and computer readable storage medium
CN111767776A (en) * 2019-12-28 2020-10-13 西安宇视信息科技有限公司 Abnormal license plate selection method and device
CN112150795A (en) * 2019-06-26 2020-12-29 杭州海康威视数字技术股份有限公司 Method and device for detecting vehicle track abnormity
CN112364820A (en) * 2020-11-27 2021-02-12 深源恒际科技有限公司 Deep learning-based vehicle insurance underwriting and vehicle checking picture acquisition method and system
CN112507957A (en) * 2020-12-21 2021-03-16 北京百度网讯科技有限公司 Vehicle association method and device, road side equipment and cloud control platform
CN112577487A (en) * 2019-09-27 2021-03-30 阿里巴巴集团控股有限公司 Route processing method and device and electronic equipment
CN112991820A (en) * 2021-04-26 2021-06-18 亿海蓝（北京）数据技术股份公司 Fake plate ship identification method and system
CN113284342A (en) * 2021-05-17 2021-08-20 上海欧菲智能车联科技有限公司 Empty parking space detection method and device, vehicle and computer equipment
CN113382361A (en) * 2020-03-09 2021-09-10 上海擎感智能科技有限公司 Route segmentation based test driving monitoring method, system, medium and equipment
CN113642893A (en) * 2021-08-16 2021-11-12 公安部交通管理科学研究所 New energy automobile operation risk assessment method
CN115083038A (en) * 2022-06-29 2022-09-20 南京理工大学泰州科技学院 Device and method for detecting vehicle driving abnormity
CN116071931A (en) * 2022-12-29 2023-05-05 北京中科神通科技有限公司 Expressway traffic vehicle information prediction method and system
WO2023088201A1 (en) * 2021-11-19 2023-05-25 杭州海康威视数字技术股份有限公司 Method, apparatus and device for detecting abnormal vehicle

Families Citing this family (26)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111133801B (en) 2017-09-28 2023-03-21 瑞典爱立信有限公司 Frequency or Radio Access Technology (RAT) selection based on slice availability
CN110415512B (en) * 2018-04-28 2021-04-27 杭州海康威视数字技术股份有限公司 Vehicle information management method, device and storage medium
CN109903045B (en) * 2019-01-24 2024-05-03 平安科技（深圳）有限公司 Behavior track monitoring method, device, computer equipment and medium
CN110162588B (en) * 2019-05-29 2023-06-09 浪潮软件集团有限公司 Track fusion method of multidimensional associated information
CN111368868B (en) * 2019-07-31 2024-01-09 杭州海康威视系统技术有限公司 Method, device and equipment for determining vehicle fake license plate
CN111078973B (en) * 2019-12-16 2023-10-20 浙江省北大信息技术高等研究院 Fake-licensed car identification method, equipment and storage medium based on big data
CN111159254B (en) * 2019-12-30 2023-07-25 武汉长江通信产业集团股份有限公司 Vehicle and personnel association method based on big data processing
CN113766426B (en) * 2020-06-29 2024-04-12 京东城市(北京)数字科技有限公司 Early warning method and early warning device for vehicle running risk, electronic equipment and storage medium
CN111899521A (en) * 2020-07-17 2020-11-06 南京师范大学泰州学院 Method for detecting vehicle violation behaviors on expressway
CN111950471B (en) * 2020-08-14 2024-02-13 杭州海康威视系统技术有限公司 Target object identification method and device
CN112200044B (en) * 2020-09-30 2024-04-30 北京四维图新科技股份有限公司 Abnormal behavior detection method and device and electronic equipment
CN112257589B (en) * 2020-10-22 2024-03-05 大连交通大学 Detection method for abnormal parking vehicle
CN112309126B (en) * 2020-10-30 2022-07-05 杭州海康威视数字技术股份有限公司 License plate detection method and device, electronic equipment and computer readable storage medium
CN112749638A (en) * 2020-12-28 2021-05-04 深兰人工智能(深圳)有限公司 Error screening method for visual recognition track and visual recognition method for sales counter
CN112818954B (en) * 2021-03-17 2022-08-26 重庆紫光华山智安科技有限公司 Vehicle state identification method, device, equipment and medium
CN112927117B (en) * 2021-03-22 2022-08-23 上海京知信息科技有限公司 Block chain-based vehicle management communication method, management system, device and medium
CN113256997B (en) * 2021-04-30 2022-09-06 贵州数据宝网络科技有限公司 Traffic vehicle violation detection device and method
CN113535819B (en) * 2021-07-15 2023-02-28 广州交信投科技股份有限公司 Traffic situation perception analysis method and device, computer storage medium and terminal
CN114328622A (en) * 2021-12-28 2022-04-12 成都路行通信息技术有限公司 Data anomaly capture real-time processing method and system for large data flow type calculation
CN114566056B (en) * 2022-02-28 2023-01-24 招商局重庆公路工程检测中心有限公司 Highway tunnel driving safety risk identification, prevention and control method and system
CN114842462B (en) * 2022-04-07 2024-04-09 深圳市捷顺科技实业股份有限公司 License plate anti-fake method
CN114999171A (en) * 2022-05-19 2022-09-02 杭州海康威视数字技术股份有限公司 Lane change monitoring processing method, device and system
CN114913690B (en) * 2022-07-13 2022-10-11 北京融信数联科技有限公司 Traffic refined monitoring method and system and readable storage medium
CN115880898B (en) * 2022-11-10 2024-05-10 公安部交通管理科学研究所 Private new energy automobile abnormal use judging method
CN116386339B (en) * 2023-06-06 2023-08-15 中路高科交通科技集团有限公司 Intelligent detection method for illegal running behavior of inter-provincial passenger transport not according to line
CN117177213B (en) * 2023-09-25 2024-02-13 北京皓宽网络科技有限公司 Data processing method and system based on data exchange

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012058851A (en) * 2010-09-06 2012-03-22 Toshiba Corp Abnormal vehicle detecting device
CN103116989A (en) * 2013-01-29 2013-05-22 江苏省邮电规划设计院有限责任公司 Smart city based fake platevehicle capture method
CN103578277A (en) * 2012-08-07 2014-02-12 上海弘视通信技术有限公司 Method and device for searching fake plate suspicion vehicle
CN103914986A (en) * 2014-03-14 2014-07-09 浙江宇视科技有限公司 Method and device for fake-license-plate analysis
CN104200669A (en) * 2014-08-18 2014-12-10 华南理工大学 Fake-licensed car recognition method and system based on Hadoop

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9293047B2 (en) * 2009-01-08 2016-03-22 GM Global Technology Operations LLC Methods and system for monitoring vehicle movement for use in evaluating possible intersection of paths between vehicle
CN101540105A (en) * 2009-04-15 2009-09-23 四川川大智胜软件股份有限公司 Fake-licensed car detection method based on number-plate identification and gridding supervision
CN101635093A (en) * 2009-05-22 2010-01-27 上海宝康电子控制工程有限公司 Identification system and method of unlicensed motor vehicles
CN102509458B (en) * 2011-10-19 2015-01-14 杭州华三通信技术有限公司 Fake-licensed vehicle identifying method and device
CN103456175B (en) * 2013-09-25 2015-04-22 武汉烽火众智数字技术有限责任公司 Accompanying vehicle real-time detection method based on vehicle registration plate recognition and meshing monitoring

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2012058851A (en) * 2010-09-06 2012-03-22 Toshiba Corp Abnormal vehicle detecting device
CN103578277A (en) * 2012-08-07 2014-02-12 上海弘视通信技术有限公司 Method and device for searching fake plate suspicion vehicle
CN103116989A (en) * 2013-01-29 2013-05-22 江苏省邮电规划设计院有限责任公司 Smart city based fake platevehicle capture method
CN103914986A (en) * 2014-03-14 2014-07-09 浙江宇视科技有限公司 Method and device for fake-license-plate analysis
CN104200669A (en) * 2014-08-18 2014-12-10 华南理工大学 Fake-licensed car recognition method and system based on Hadoop

Cited By (44)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107622667A (en) * 2017-09-28 2018-01-23 广州烽火众智数字技术有限公司 A kind of adulterium number-plate number detection method and system
CN108022427A (en) * 2017-10-30 2018-05-11 深圳市赛亿科技开发有限公司 A kind of recognition methods of fake license plate vehicle and system
CN109831744B (en) * 2017-11-23 2021-05-07 腾讯科技（深圳）有限公司 Abnormal track identification method and device and storage equipment
CN109831744A (en) * 2017-11-23 2019-05-31 腾讯科技（深圳）有限公司 It is a kind of exception track recognizing method, device and storage equipment
CN108280435A (en) * 2018-01-25 2018-07-13 盛视科技股份有限公司 A kind of passenger's abnormal behaviour recognition methods based on human body attitude estimation
CN108986474A (en) * 2018-08-01 2018-12-11 平安科技（深圳）有限公司 Fix duty method, apparatus, computer equipment and the computer storage medium of traffic accident
CN108986478A (en) * 2018-09-17 2018-12-11 公安部交通管理科学研究所 A method of fixed point screens illegal vehicle
CN109637122A (en) * 2018-10-19 2019-04-16 天津易华录信息技术有限公司 A kind of data processing method, equipment and system
CN109767615A (en) * 2018-10-19 2019-05-17 江苏智通交通科技有限公司 Road network traffic flow key flow direction and critical path analysis method
CN109767615B (en) * 2018-10-19 2021-05-18 江苏智通交通科技有限公司 Method for analyzing key flow direction and key path of road network traffic flow
CN111179603A (en) * 2018-11-09 2020-05-19 杭州海康威视数字技术股份有限公司 Vehicle identification method and device, electronic equipment and storage medium
CN109657621A (en) * 2018-12-21 2019-04-19 杨文鑫 Big data trajectory analysis platform
CN109859492A (en) * 2019-03-27 2019-06-07 成都市公安科学技术研究所 A kind of system and method identifying fake-licensed car
CN112150795B (en) * 2019-06-26 2022-05-06 杭州海康威视数字技术股份有限公司 Method and device for detecting vehicle track abnormity
CN112150795A (en) * 2019-06-26 2020-12-29 杭州海康威视数字技术股份有限公司 Method and device for detecting vehicle track abnormity
CN111367906A (en) * 2019-07-23 2020-07-03 杭州海康威视系统技术有限公司 Abnormal vehicle identification method, device, equipment and computer readable storage medium
CN111367906B (en) * 2019-07-23 2023-09-05 杭州海康威视系统技术有限公司 Abnormal vehicle identification method, device, equipment and computer readable storage medium
CN112577487A (en) * 2019-09-27 2021-03-30 阿里巴巴集团控股有限公司 Route processing method and device and electronic equipment
CN110728842B (en) * 2019-10-23 2021-10-08 江苏智通交通科技有限公司 Abnormal driving early warning method based on reasonable driving range of vehicles at intersection
CN110728842A (en) * 2019-10-23 2020-01-24 江苏智通交通科技有限公司 Abnormal driving early warning method based on reasonable driving range of vehicles at intersection
CN110807694A (en) * 2019-11-06 2020-02-18 广东泰骅科技信息服务有限公司 Talent sharing system
CN110874355A (en) * 2019-11-28 2020-03-10 北京以萨技术股份有限公司 Method, system, terminal and medium for detecting abnormal behavior of vehicle loitering winding
CN110874355B (en) * 2019-11-28 2022-08-23 以萨技术股份有限公司 Method, system, terminal and medium for detecting abnormal behavior of vehicle loitering winding
CN110942640A (en) * 2019-12-04 2020-03-31 无锡华通智能交通技术开发有限公司 Method for actively discovering suspect vehicle illegally engaged in network car booking passenger transportation
CN111767776A (en) * 2019-12-28 2020-10-13 西安宇视信息科技有限公司 Abnormal license plate selection method and device
CN111767776B (en) * 2019-12-28 2024-03-29 西安宇视信息科技有限公司 Abnormal license plate selecting method and device
CN111191949A (en) * 2020-01-03 2020-05-22 北京三快在线科技有限公司 Method and device for identifying abnormal driving behaviors of networked vehicles and electronic equipment
CN111369805A (en) * 2020-01-09 2020-07-03 杭州海康威视系统技术有限公司 Fake plate detection method and device, electronic equipment and computer readable storage medium
CN111369805B (en) * 2020-01-09 2021-08-06 杭州海康威视系统技术有限公司 Fake plate detection method and device, electronic equipment and computer readable storage medium
CN111291216B (en) * 2020-02-28 2022-06-14 罗普特科技集团股份有限公司 Method and system for analyzing foothold based on face structured data
CN111291216A (en) * 2020-02-28 2020-06-16 罗普特科技集团股份有限公司 Method and system for analyzing foothold based on face structured data
CN113382361B (en) * 2020-03-09 2022-11-15 上海擎感智能科技有限公司 Route segmentation based test driving monitoring method, system, medium and equipment
CN113382361A (en) * 2020-03-09 2021-09-10 上海擎感智能科技有限公司 Route segmentation based test driving monitoring method, system, medium and equipment
CN112364820A (en) * 2020-11-27 2021-02-12 深源恒际科技有限公司 Deep learning-based vehicle insurance underwriting and vehicle checking picture acquisition method and system
CN112507957A (en) * 2020-12-21 2021-03-16 北京百度网讯科技有限公司 Vehicle association method and device, road side equipment and cloud control platform
CN112507957B (en) * 2020-12-21 2023-12-15 阿波罗智联(北京)科技有限公司 Vehicle association method and device, road side equipment and cloud control platform
CN112991820A (en) * 2021-04-26 2021-06-18 亿海蓝（北京）数据技术股份公司 Fake plate ship identification method and system
CN113284342A (en) * 2021-05-17 2021-08-20 上海欧菲智能车联科技有限公司 Empty parking space detection method and device, vehicle and computer equipment
CN113642893A (en) * 2021-08-16 2021-11-12 公安部交通管理科学研究所 New energy automobile operation risk assessment method
WO2023088201A1 (en) * 2021-11-19 2023-05-25 杭州海康威视数字技术股份有限公司 Method, apparatus and device for detecting abnormal vehicle
CN115083038A (en) * 2022-06-29 2022-09-20 南京理工大学泰州科技学院 Device and method for detecting vehicle driving abnormity
CN115083038B (en) * 2022-06-29 2024-02-27 南京理工大学泰州科技学院 Device and method for detecting abnormal running of vehicle
CN116071931A (en) * 2022-12-29 2023-05-05 北京中科神通科技有限公司 Expressway traffic vehicle information prediction method and system
CN116071931B (en) * 2022-12-29 2024-01-09 北京中科神通科技有限公司 Expressway traffic vehicle information prediction method and system

Also Published As

Publication number Publication date
WO2017157119A1 (en) 2017-09-21

Similar Documents

Publication Publication Date Title
CN107204114A (en) A kind of recognition methods of vehicle abnormality behavior and device
CN109448370B (en) Traffic control subarea division method based on vehicle track data
CN106846801A (en) A kind of region based on track of vehicle is hovered anomaly detection method
CN111680613B (en) Method for detecting falling behavior of escalator passengers in real time
CN108734129A (en) mobile phone and vehicle location analysis method and system
CN106314438A (en) Method and system for detecting abnormal track in driver driving track
CN102254428B (en) Traffic jam detection method based on video processing
CN110288823B (en) Traffic violation misjudgment identification method based on naive Bayesian network
CN109840523A (en) A kind of municipal rail train Train number recognition algorithm based on image procossing
CN109344886B (en) Occlusion number plate distinguishing method based on convolutional neural network
CN106373461A (en) Driver training management service system and method
CN104021677A (en) Method and device for detecting turnover of number plate
CN116168356B (en) Vehicle damage judging method based on computer vision
CN111444843B (en) Multimode driver and vehicle illegal behavior monitoring method and system
CN110164137B (en) Method, system and medium for identifying fake-licensed vehicle based on driving time of bayonet pair
Hamedmoghadam et al. Automated extraction of origin-destination demand for public transportation from smartcard data with pattern recognition
CN114494998B (en) Intelligent analysis method and system for vehicle data
CN114067610B (en) Simulation scene construction method and device for missing turnout accident
CN111524350B (en) Method, system, terminal device and medium for detecting abnormal driving condition of vehicle and road cooperation
CN108171187A (en) A kind of abnormal behaviour automatic identifying method and device based on the extraction of bone point
CN116778460A (en) Fatigue driving identification method based on image identification
CN114023085B (en) Intersection signal timing parameter inference method based on bayonet detection data
CN107423758A (en) Abstraction sequence vital point drives training cheating class hour recognition methods from trend point
CN111126144B (en) Vehicle track abnormity detection method based on machine learning
CN107145835A (en) Driver's cutting load testing car-mounted device based on image recognition

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
WW01 Invention patent application withdrawn after publication

Application publication date: 20170926