Specific embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on this
Embodiment in invention, the every other reality that those of ordinary skill in the art are obtained under the premise of not making creative work
Apply example, broadly fall into the scope of protection of the invention.
The flow process that Fig. 1 show a kind of vehicle peccancy line ball intelligent control method of one embodiment of the invention offer is illustrated
Figure.As shown in figure 1, this vehicle peccancy line ball intelligent control method includes:
Step 101:When identifying that in the video monitoring data of Real-time Collection vehicle is blinded by roadmarking, intercepting should
The present frame monitoring data of video monitoring data.
Specifically, the identification process for vehicle defiladed route graticule behavior can be accomplished in several ways.For example, may be used
Vehicle Object and roadmarking object are first extracted from video monitoring data based on existing image object extractive technique, when this
When Vehicle Object and this roadmarking object cross, then illustrate that this vehicle is blinded by this roadmarking.But the present invention couple
The identification process of vehicle defiladed route graticule behavior is simultaneously not specifically limited.
In the prior art, the video monitoring data of Real-time Collection is usually stored, is needing to carry out video to divide
Again the video monitoring data of storage is recalled or by manually carrying out discriminatory analysis during analysis, thereby resulted in hysteresis effect it is impossible to
Analyze the timely process realized under scene to vehicle peccancy line ball behavior in vehicle monitoring.And in embodiments of the present invention, when
When identifying that vehicle is blinded by roadmarking in the video monitoring data of Real-time Collection, just intercept present frame monitoring data, deserve
Previous frame monitoring data is directly as the basic resources data subsequently carrying out further real-time intelligent analysis.
In an embodiment of the present invention, this present frame monitoring data may include in present frame monitoring image and following items
One or more:Video acquisition positional information and video acquisition temporal information.Wherein present frame monitoring image is subsequently to carry out
Roadmarking information and the resource data of traffic mark information extraction that vehicle characteristic information, vehicle are covered, video acquisition position
Confidence breath and video acquisition temporal information can as additional attribute information be used for that clear and definite vehicle peccancy line ball behavior occurs when
Between and place.
In an embodiment of the present invention, present frame monitoring data also can directly be issued the monitoring personnel of Surveillance center and check,
Enable automatically carrying out of monitoring personnel manual oversight whole vehicle peccancy line ball intelligent monitoring process.
Step 102:It is blinded by the vehicle characteristic information of vehicle, the car of roadmarking in extract real-time present frame monitoring data
The roadmarking information covered and traffic mark information.
In an embodiment of the present invention, vehicle characteristic information may include one or more of following items:Vehicle identification number,
The time span that vehicle heading, Vehicle Speed and vehicle line ball travel, for clear and definite vehicle further feature with
And the travelling characteristic of vehicle.The roadmarking information that vehicle is covered includes:The classification information of roadmarking, such as bus are special
With lane line, track solid line, channelizing line, stop line and zebra stripes etc..Traffic mark information may include traffic lights information, is used for joining
Close the traffic rules that roadmarking information is suitable for clear and definite roadmarking.
In an embodiment of the present invention, this vehicle characteristic information, vehicle are covered roadmarking information and traffic mark
Roadmarking information and the pre- instruction of traffic mark information that the extraction process of information can be covered based on vehicle characteristic information, vehicle
Practice model realization.For example, in advance pre-training is carried out to the corresponding relation of type of vehicle and the shape of vehicle, to roadmarking and road
The corresponding relation of the shape facility of marking lines carries out pre-training, when the build of the vehicle identifying in monitor video is less and without
When having wagon box, can determine whether that this vehicle is car based on pre-training model;There is " public transport is special " etc. when identifying in roadmarking
During printed words, can determine whether that this roadmarking is bus special-purpose vehicle diatom based on pre-training model.
Step 103:The roadmarking information covered according to the vehicle characteristic information extracting, vehicle and traffic mark information
Judge that this vehicle whether there is line ball behavior violating the regulations.
Can confirm that the feature of vehicle based on this vehicle characteristic information, can be true based on the roadmarking information that this vehicle is covered
Recognize the roadmarking that vehicle is covered, the traffic rules that this roadmarking is suitable for be can confirm that based on this traffic mark information, when
The roadmarking that the feature of vehicle and vehicle are covered does not correspond, or the feature of the roadmarking that covered of vehicle and traffic mark
Just can assert when not corresponding that vehicle has line ball behavior violating the regulations, it is achieved thereby that to the vehicle peccancy line ball behavior in monitor video
Real-time analysis judge.In an embodiment of the present invention, after judging that vehicle has line ball behavior violating the regulations, can be based further on
The roadmarking information that vehicle characteristic information, vehicle are covered and traffic mark information provide punishment on contravention of regulation result.Such as " vehicle
A there are the illegal activities of line ball traveling it is proposed that deduction XX divides, fine YY unit ... ".
In an embodiment of the present invention, can be pre-training model based on the machine description to clause violating the regulations, sentence in real time
Whether disconnected vehicle is illegal to be blinded by roadmarking.For example, by the limited traveling of the different roadmarking of pre-training procedural learning
Type of vehicle, such as only allow in bus special-purpose vehicle diatom bus travel.Vehicle included by when vehicle characteristic information
Type is car, and when the roadmarking type included by roadmarking information is bus special-purpose vehicle diatom, when this little vapour
Car just can determine whether when being pressed onto this bus special-purpose vehicle diatom that this car has line ball behavior violating the regulations.Again for example, can be instructed by pre-
Practice the traffic rules that procedural learning difference roadmarking is suitable for, such as the roadmarking type included by when roadmarking information
For zebra stripes, when the traffic lights information included by traffic mark information is green light, if the vehicle speed included by vehicle characteristic information
Degree is zero, then illustrate that this vehicle stops on zebra stripes, there is the behavior of masking zebra stripes violating the regulations.
In an embodiment of the present invention, the vehicle characteristic information of vehicle, vehicle are covered roadmarking information and friendship
Logical identification information can be conserved after being extracted, and reviewing and inquiring about in order to follow-up.For example, can receive with vehicle spy
After reference breath, the roadmarking information covered of vehicle or query statement that traffic mark information is querying condition, transfer and car
The corresponding all roadmarking information of characteristic information and traffic mark information, or transfer corresponding with roadmarking information all
Vehicle characteristic information and traffic mark information, or transfer all vehicle characteristic information corresponding with traffic mark information and road road sign
Line information, thus realize the line ball vehicle violating the regulations of the line ball record violating the regulations for a certain vehicle, a certain concrete roadmarking
The inquiry of the line ball record violating the regulations under record or a certain traffic mark.For example, if receiving with bus special lane graticule B for looking into
During the query statement of inquiry condition, just may bring up all cars covering this bus special lane graticule B in current video monitoring data
Characteristic information and this bus special lane graticule B near traffic mark information.
In an embodiment of the present invention, because the normally travel behavior of vehicle does not need to carry out process violating the regulations, if therefore
It is judged as that vehicle does not have line ball behavior violating the regulations, then can not preserve this vehicle characteristic information, roadmarking that vehicle is covered letter
Breath and traffic mark information, to improve the utilization ratio of storage hardware resource.
The flow process that Fig. 2 show a kind of vehicle peccancy line ball intelligent control method of another embodiment of the present invention offer is illustrated
Figure.As shown in Fig. 2 this vehicle peccancy line ball intelligent control method includes:
Step 201:When identifying that in the video monitoring data of Real-time Collection vehicle is blinded by roadmarking, intercepting should
The present frame monitoring data of video monitoring data.
Step 202:The area video monitoring data in region residing for vehicle is intercepted from present frame monitoring data.
By intercepting area video monitoring data from present frame monitoring data, eliminate unrelated with current monitor vehicle
The video monitoring data in region, decreases subsequent extracted vehicle characteristic information, roadmarking information that vehicle is covered and traffic
The amount of calculation of identification information, mitigates the computation burden of hardware analysis resource.
It should be appreciated that the area size of this area video monitoring data can be by developer according to actual scene demand
Fixed, the monitoring range of such as head end video collecting device includes two blocks, and monitoring personnel of interest be only therein one
The line ball situation violating the regulations of individual block, now just can intercept from video monitoring data and monitor number when the area video of previous block
According to.Or, depending on the size in this region also can be by the image size of the vehicle being monitored.However, the present invention is to this area video
The determination mode of the area size of monitoring data does not limit.
Step 203:The roadmarking information that extraction vehicle characteristic information, vehicle are covered from area video monitoring data
With traffic mark information.
Because original video monitoring data has been intercepted for area video monitoring data, therefore vehicle characteristics letter
The amount of calculation of the extraction process of breath, the roadmarking information that vehicle is covered and traffic mark information is reduced, and alleviates
The computation burden of hardware analysis resource.Roadmarking information that vehicle characteristic information, vehicle are covered and traffic mark information
Concrete extracting mode still can pass through pre-training model realization, will not be described here.
In an embodiment of the present invention it is contemplated that the extraction of roadmarking information covered of vehicle characteristic information and vehicle
It is concerned only with the vehicle in monitor video, the traffic mark in monitor video is paid close attention in the extraction of traffic mark information, and vehicle is in prison
In control video, shared region is less than the region shared by traffic mark, therefore can also only carry from area video monitoring data
Take the roadmarking information that vehicle characteristic information and vehicle are covered, and traffic mark information is still from the video monitoring of acquired original
Extracting data.However, present invention roadmarking information that vehicle characteristic information, vehicle are covered and traffic mark information
Extract basis and be not specifically limited.
Step 204:Roadmarking information and traffic mark that vehicle characteristic information based on extract real-time, vehicle are covered
Information generative semantics analysis result.
For example, the characteristic information of vehicle includes vehicle A is large truck, and travel speed is 100km/h, and vehicle is covered
Roadmarking information is runway graticule B, and traffic mark information is speed limit 80km/h, then the semantic analysis result being generated can table
It is shown as the runway graticule B that large truck A is blinded by speed limit 80km/h with the travel speed of 100km/h.
Step 205:Based on the pre-training model of semantic analysis result and the machine description of clause violating the regulations, real-time judge institute
State and be blinded by the vehicle of roadmarking and whether illegal be blinded by roadmarking.
It should be noted that while for purposes of simplicity of explanation method of the present invention is illustrated and described as a succession of dynamic
Make, it should be understood that and recognizing that the not execution sequence institute by these actions is limited by claimed subject content, because one
A little actions concurrently can be occurred according to being occurred with order different shown and described herein or with other actions, with
When some actions be also possible that some sub-steps, and between these sub-steps, be likely to occur the possibility that execution is intersected on sequential.
For example, in an embodiment of the present invention, as shown in Fig. 2 ', identify that vehicle hides in the video monitoring data of Real-time Collection working as
When having covered roadmarking, and after intercepting the present frame monitoring data of this video monitoring data (step 201 '), traffic mark information
Can directly extract from present frame monitoring data (step 202 '), then intercept area residing for vehicle again from present frame monitoring data
The area video monitoring data (step 203 ') in domain and extract vehicle characteristic information and vehicle from area video monitoring data
The roadmarking information (step 204 ') covered, afterwards based on the traffic mark information extracted, vehicle characteristic information and
Roadmarking information generative semantics analysis result that vehicle is covered (step 205 '), and be based ultimately upon this semantic analysis result with
And the pre-training model of the machine description of clause violating the regulations, whether real-time judge vehicle is illegal to be blinded by roadmarking (step
206’).Additionally, it is possible to simultaneously the action of not all diagram is necessary to the method implemented according to appended claims.
Furthermore, the description of abovementioned steps is not excluded for the method and can also include obtaining the additional step of additional effect.Also should manage
Solve, the method and step described in different embodiments or flow process can be mutually combined or replace.
Fig. 3 show a kind of structural representation of vehicle peccancy line ball intelligent monitor system of one embodiment of the invention offer
Figure.As shown in figure 3, this vehicle peccancy line ball intelligent monitor system 30 includes:
Environmental perception device 301, is configured to work as and identifies that in the video monitoring data of Real-time Collection vehicle is blinded by
During marking lines, intercept the present frame monitoring data of described video monitoring data;
Characteristics of objects analytical equipment 302, is configured to monitor number according to the present frame that described environmental perception device 301 is gathered
According to extract real-time is blinded by the roadmarking information that the vehicle characteristic information of the vehicle of roadmarking and vehicle are covered;
Geographical sign analytical equipment 303, is configured to the present frame monitoring data being gathered according to environmental perception device 301,
Extract real-time traffic mark information;And
Decision making device 304, is configured to the vehicle characteristic information according to characteristics of objects analytical equipment 302 extraction and vehicle is hidden
The roadmarking information covered, and geographical sign analytical equipment 303 extract traffic mark information judge whether described vehicle is deposited
In line ball behavior violating the regulations.
In an embodiment of the present invention, this vehicle peccancy line ball intelligent monitor system further includes:Semantic analysis device
305, it is configured to described characteristics of objects analytical equipment 302 and the distinguished extract real-time of described geographical sign analytical equipment 303
Vehicle characteristic information, the roadmarking information covered of vehicle and traffic mark information generative semantics analysis result;
Wherein decision making device 304 is further configured to:
The semantic analysis result being generated based on semantic analysis device 305 and the pre-training of the machine description of clause violating the regulations
Model, the vehicle being blinded by roadmarking described in real-time judge whether illegal is blinded by roadmarking.
In an embodiment of the present invention, this decision making device 304 is further configured to:
Receive the roadmarking information covered with vehicle characteristic information, vehicle or traffic mark information as querying condition
Query statement;And transfer all roadmarking information corresponding with vehicle characteristic information and traffic mark information, or transfer with
The corresponding all vehicle characteristic information of roadmarking information and traffic mark information, or transfer institute corresponding with traffic mark information
There are vehicle characteristic information and roadmarking information.
In an embodiment of the present invention, decision making device 304 is further configured to:
If vehicle does not have line ball behavior violating the regulations, do not preserve vehicle characteristic information corresponding with vehicle, vehicle is covered
Roadmarking information and traffic mark information.
In an embodiment of the present invention, this vehicle peccancy line ball intelligent monitor system further includes:
Subject area extraction element 306, is configured to be blinded by roadmarking described in intercepting from present frame monitoring data
The area video monitoring data in region residing for vehicle;
Wherein characteristics of objects analytical equipment 302 is further configured to:Vehicle characteristics are extracted from area video monitoring data
The roadmarking information that information and/or vehicle are covered.
In an alternative embodiment of the invention, geographical sign analytical equipment 303 also can be further configured to:From area video prison
Control extracting data traffic mark information.
In an embodiment of the present invention, characteristics of objects analytical equipment 302 is further configured to:Based on to vehicle characteristic information
With the pre-training model of roadmarking information, extract vehicle characteristic information and roadmarking information from present frame monitoring data;
And/or,
Geographical sign analytical equipment 303 is further configured to:Based on the pre-training model of traffic mark information, from present frame
Traffic mark information is extracted in monitoring data.
In an embodiment of the present invention, this decision making device 304 is further configured to:When judged result is to be, according to carrying
The roadmarking information that the vehicle characteristic information that takes, vehicle are covered and traffic mark information provide punishment on contravention of regulation result.
As can be seen here, the vehicle peccancy line ball intelligent monitor system that the embodiment of the present invention is provided is based on a kind of prison in fact
Control video intelligent information model is realized.Information flow in this monitor video intelligent information model can be divided into different aspects to carry
Take, and there is certain dependence between adjacent aspect, as shown in Figure 4.Intercept present frame in environmental perception device 301
During video monitoring data, by present frame video with live perception data (such as:Sound, the time, the geographical position of camera,
Temperature, meteorology, camera pose etc.) it is stored in environment sensing layer, these information are traditional video surveillance and intelligent video monitoring provides
The fundamental of monitoring scene, provides necessary support for top layer decision-making.During front-end processing, by present frame video monitoring
Data carries out preliminary treatment (include conventional pretreatment, and front end intellectual analysis based on statistical learning method etc.), and will be just
The result that step is processed is stored in characteristic layer, geographical sign layer and object layer, and corresponding objects feature analyzing apparatus 302 extract vehicle respectively
Vehicle characteristic information and vehicle covered the process of roadmarking information, geographical sign analytical equipment 303 extract traffic mark
The process of knowledge information and subject area extraction element 306 intercept area video monitoring data from present frame video monitoring data
Process.In back-end processing procedures, by according to different application needs, comprehensive aforesaid accordingly each layer, using machine learning
Technology is analyzed and processes, and related result is stored in semantic layer, corresponding semantic analysis device 305 generative semantics analysis
The process of result.For the semantic analysis result in semantic analysis device 305, using the judgment models of machine learning techniques training
Different decision recommendations can be given and supply monitoring personnel reference;Meanwhile, monitoring personnel can also send instruction to system, to inquire about
Corresponding content in monitoring data, these belong to decision-making/understand layer.Judgment models therein and by its judge to obtain certainly
Plan is advised, belongs to decision-making category.Monitoring personnel, during observing, sends certain instruction to inquire about a certain event to monitoring system
In, possess the target of certain category feature, system is instructed the describing mode being construed to meet this structural model, and it is grasped
Data enters line retrieval, belongs to and understands category.
It should be appreciated that when vehicle peccancy line ball intelligent monitor system includes head end video collecting device and rear end video analysis
During equipment, environmental perception device 301 may be provided in head end video collecting device, and geographical sign analytical equipment 303, target area
Domain extraction element 306, characteristics of objects analytical equipment 302, semantic analysis device 305, decision making device 304 can be separately positioned on front end
In video capture device or be arranged in the video analysis equipment of rear end.As long as all in vehicle peccancy line ball intelligent monitor system
Device is capable of the progressively extraction of respective analysis abstraction function and information flow, with the purpose being finally reached semantic decision-making is
Can.The present invention is specifically located at head end video collecting device or rear end to the device in vehicle peccancy line ball intelligent monitor system
Video analysis equipment does not limit.
The teachings of the present invention is also implemented as a kind of computer program of computer-readable recording medium, including meter
Calculation machine program code, when computer program code is by computing device, it enables a processor to according to embodiment party of the present invention
The method of formula is realizing vehicle peccancy line ball intelligent control method as the embodiment described herein.Computer-readable storage medium is permissible
For any tangible media, such as floppy disk, CD-ROM, DVD, hard disk drive, even network medium etc..
It should be understood that although a kind of way of realization the foregoing describing embodiment of the present invention can be computer program product
Product, but the method or apparatus of embodiments of the present invention can be come in fact by the combination according to software, hardware or software and hardware
Existing.Hardware components can be realized using special logic;Software section can store in memory, by suitable instruction execution
System, such as microprocessor or special designs hardware are executing.It will be understood by those skilled in the art that above-mentioned side
Method and equipment can be realized in processor control routine using computer executable instructions and/or be included in, such as such as
Disk, the mounting medium of CD or DVD-ROM, such as the programmable memory of read-only storage (firmware) or such as optics or
Such code is provided on the data medium of electrical signal carrier.Methods and apparatus of the present invention can be by such as ultra-large
The semiconductor of integrated circuit OR gate array, logic chip, transistor etc. or such as field programmable gate array, can compile
The hardware circuit of the programmable hardware device of journey logical device etc. is realized it is also possible to soft by various types of computing devices
Part is realized it is also possible to be realized by the combination such as firmware of above-mentioned hardware circuit and software.
It will be appreciated that though be referred to some modules or the unit of device in detailed descriptions above, but this stroke
Divide and be merely exemplary rather than enforceable.In fact, according to an illustrative embodiment of the invention, above-described two or
The more feature of multimode/unit and function can be realized in a module/unit, conversely, an above-described module/mono-
The feature of unit and function can be to be realized by multiple module/units with Further Division.Additionally, above-described certain module/
Unit can be omitted under some application scenarios.For example when the computing capability of hardware computing resource is unrestricted, also can not wrap
Include subject area extraction element.
It is also understood that in order to not obscure embodiments of the present invention, specification only to some are crucial, may not necessary skill
Art and feature are described, and the feature that may some those skilled in the art be capable of is explained.
The foregoing is only presently preferred embodiments of the present invention, not in order to limit the present invention, all essences in the present invention
Within god and principle, any modification of being made, equivalent etc., should be included within the scope of the present invention.