CN104809874B - A kind of traffic incidents detection method and apparatus - Google Patents
A kind of traffic incidents detection method and apparatus Download PDFInfo
- Publication number
- CN104809874B CN104809874B CN201510182223.9A CN201510182223A CN104809874B CN 104809874 B CN104809874 B CN 104809874B CN 201510182223 A CN201510182223 A CN 201510182223A CN 104809874 B CN104809874 B CN 104809874B
- Authority
- CN
- China
- Prior art keywords
- model
- vehicle
- track
- traffic
- car
- 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.)
- Active
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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Abstract
The invention provides a kind of detection method of traffic events and device, methods described includes：Obtain Traffic Surveillance Video image of the video camera in gathering when presetting presetting bit；Track of vehicle model and/or car week environmental model are set up according to the Traffic Surveillance Video image；According to the track of vehicle model and the pavement structure model inspection traffic events for pre-building, and/or, according to car week environmental model and the pavement structure model inspection traffic events for pre-building；The pavement structure model is the model that is set up according to the monitor video image that traffic events do not occur on road surface.This method by traffic incident detecting system automatic detection traffic events saves substantial amounts of human and material resources, reduces the working strength of monitoring personnel.And will not occur due to the deficient in energy risk for causing the wrong report to traffic events and failing to report, therefore, the detection method improves the accuracy of traffic incidents detection.
Description
Technical field
The present invention relates to road monitoring field, more particularly to a kind of traffic incidents detection method and apparatus.
Background technology
Preventing road monitoring system is played an important role in public security prevention and control.In order to obtain the traffic of road in real time,
Preventing road monitoring system is provided with some Important Sections, road real time status are shot by video camera, the figure that then will be shot
As uploading to road monitoring command centre in real time.
In existing preventing road monitoring system, the monitoring personnel of road monitoring command centre is uploaded by operating head observation
Road all directions monitor video image, artificially judge the generation of traffic events by observing monitor video image.This
Plant traffic event monitoring method and consume substantial amounts of human and material resources resource.And as the energy of monitoring personnel is limited, it is easy to
Cause the wrong report of traffic events and fail to report.
Content of the invention
In view of this, the invention provides a kind of traffic incidents detection method and apparatus, with realize to traffic events from
Dynamic detection.
In order to above-mentioned technical problem is solved, following technical scheme is present invention employs：
A kind of detection method of traffic events, including：
Obtain Traffic Surveillance Video image of the video camera in gathering when presetting presetting bit；
Track of vehicle model and/or car week environmental model are set up according to the Traffic Surveillance Video image；
According to the track of vehicle model and the pavement structure model inspection traffic events for pre-building, and/or, according to institute
State car week environmental model and the pavement structure model inspection traffic events for pre-building；The pavement structure model be according to road surface
On do not occur traffic events monitor video image set up model.
Alternatively, the traffic incidents detection method is carried out automatically between the presetting bit of multiple differences using every road video camera
The method of taking turn monitoring, can cover the cross section on whole road surface per the multiple differences presetting bit residing for the video camera of road, and setting is taken the photograph
After the monitoring period of a presetting bit reaches Preset Time, automatic alternate is monitored camera to next presetting bit.
It is alternatively, described according to the track of vehicle model and the pavement structure model inspection traffic events that pre-build,
And/or, after car week environmental model and the pavement structure model inspection traffic events for pre-building, also include：
After the instruction for carrying out manual intervention is received, cancel and presetting bit associated profile is loaded, receiving certainly is carried out
The instruction of manual intervention is risen to reaching after default manual monitoring duration, points out whether load presetting bit associated profile, if
It is to control camera switching to the position of the next presetting bit of the default presetting bit, to gather the default presetting bit
Traffic on the corresponding road of next presetting bit, if not, extend artificial monitor duration, receives that prolongation is artificial to be supervised certainly
The instruction of control duration is risen to being reached after default manual monitoring duration again, is returned and is executed whether the prompting loads presetting bit correlation
The step of configuration file；
After the instruction that refusal carries out manual intervention is received, judge video camera current presetting bit monitor event whether
Preset Time is reached, if it is, next presetting bit of the control camera switching to the default presetting bit, and return execution institute
The step of stating Traffic Surveillance Video image of the acquisition video camera in gathering when presetting presetting bit.
Alternatively, described track of vehicle model is set up according to the Traffic Surveillance Video image include：
Detect the vehicle in the Traffic Surveillance Video image；
Vehicle to detecting is tracked, and obtains the movement locus of the vehicle；
According to the movement locus of the vehicle, track of vehicle model is set up.
Alternatively, the track of vehicle model and the pavement structure model all include vehicle motion direction angle model and
Velicle motion velocity model；
Described specifically included according to track of vehicle model and the pavement structure model inspection traffic events that pre-build：
Relatively the direction angle model in the track of vehicle model and the pavement structure model and/or rate pattern, obtain
Direction angular displacement and/or velocity deviation to the track of vehicle model and the pavement structure model；
Judge whether the direction angular displacement reaches first pre-conditioned, and/or, judge whether the velocity deviation reaches
Second is pre-conditioned, to judge whether traffic events occur on road surface；
Inclined when the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model
When difference is more than angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is big
When First Speed threshold value, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is big
When second speed threshold value, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, sentence
Surely there is vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed
During threshold value, judge vehicle congestion event；
When the time that the track of vehicle model of at least two vehicles on road surface overlap occurs and overlaps is continued above second
Between section when, judge occur vehicle crash event.
Alternatively, set up car week environmental model according to the Traffic Surveillance Video image to specifically include：
Vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image；
Car week environment contour feature figure is obtained according to the vehicle and surrounding environment part；
Car week environment DPM model is set up according to car week environment contour feature figure, the car week environment DPM model includes
Root model p0, the first partial model p1, the second partial model p2, the 3rd partial model p3With the 4th partial model p4, described mould
Type p0For auto model, the first partial model p1, the second partial model p2, the 3rd partial model p3With the 4th partial model p4
Environmental model respectively on the four direction all around of vehicle.
Alternatively, described have according to car week environmental model and the pavement structure model inspection traffic events for pre-building
Body includes：
Vehicle environmental model is mated with the pavement structure model, is obtained matching value；
When matching value is more than predetermined threshold value, from all environment of car, it is partitioned into car Zhou Yichang object；
Movement locus to the car Zhou Yichang object is tracked, and sets up the locus model of car Zhou Yichang object；
When the medium velocity of the locus model of car Zhou Yichang object is 0, start timing, when the stop of car Zhou Yichang object
Between reach Preset Time after, determine the car Zhou Yichang object for shedding thing.
A kind of detection means of traffic events, including：
Acquiring unit, for obtaining the Traffic Surveillance Video image gathered when video camera is in and presets presetting bit；
Unit set up by model, for setting up track of vehicle model and/or car chow ring according to the Traffic Surveillance Video image
Border model；
Detector unit, for according to the track of vehicle model and the pavement structure model inspection traffic thing for pre-building
Part, and/or, according to car week environmental model and the pavement structure model inspection traffic events for pre-building；The road surface knot
Structure model is the model that basis does not occur that monitor video image during traffic events is set up.
Alternatively, the traffic incidents detection device is carried out automatically between the presetting bit of multiple differences using every road video camera
The method of taking turn monitoring carries out the detection of traffic events, can cover whole road per the multiple differences presetting bit residing for the video camera of road
The cross section in face, sets video camera after the monitoring period of a presetting bit reaches Preset Time, and automatic alternate is pre- to the next one
Set is monitored.
Alternatively, the model is set up unit and is included：
Detection sub-unit, for detecting the vehicle in the Traffic Surveillance Video image；
Tracking subelement, for being tracked to the vehicle for detecting, obtains the movement locus of the vehicle；
Subelement set up by track of vehicle model, for the movement locus according to the vehicle, sets up track of vehicle model.
Alternatively, the track of vehicle model and the pavement structure model all include vehicle motion direction angle model and
Velicle motion velocity model；
The detector unit includes：
Comparing subunit, for the direction angle model in relatively the track of vehicle model and the pavement structure model
And/or rate pattern, direction angular displacement and/or the speed for obtaining the track of vehicle model and the pavement structure model is inclined
Difference；
Judgment sub-unit, first pre-conditioned for judging whether the direction angular displacement reaches, and/or, the speed
It is second pre-conditioned whether deviation reaches, to judge whether traffic events occur on road surface；
Inclined when the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model
When difference is more than angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is big
When First Speed threshold value, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is big
When second speed threshold value, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, sentence
Surely there is vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed
During threshold value, judge vehicle congestion event；
When the time that the track of vehicle model of at least two vehicles on road surface overlap occurs and overlaps is continued above second
Between section when, judge occur vehicle crash event.
Alternatively, the model is set up unit and is included：
Subelement is chosen, for vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image；
Subelement is obtained, for car week environment contour feature figure being obtained according to the vehicle and surrounding environment part；
Subelement set up by all environment DPM models of car, for setting up car week environment according to car week environment contour feature figure
DPM model, the car week environment DPM model include root model p0, the first partial model p1, the second partial model p2, the 3rd local
Model p3With the 4th partial model p4, described model p0For auto model, the first partial model p1, the second partial model
p2, the 3rd partial model p3With the 4th partial model p4Environmental model respectively on the four direction all around of vehicle.
Alternatively, the detector unit includes：
Coupling subelement, for being mated vehicle environmental model with the pavement structure model, obtains matching value；
Segmentation subelement, for when the matching value is more than predetermined threshold value, being partitioned into car Zhou Yichang from all environment of car
Object；
Car Zhou Yichang object locus model sets up subelement, for the movement locus of the car Zhou Yichang object is carried out with
Track, sets up the locus model of car Zhou Yichang object；
Judge shed thing subelement, for the locus model when car Zhou Yichang object medium velocity be 0 when, start timing,
After the time of staying of car Zhou Yichang object Preset Time is reached, determine the car Zhou Yichang object for shedding thing.
Compared to prior art, the invention has the advantages that：
In the detection method of the traffic events that the present invention is provided, traffic incident detecting system can be according to track of vehicle model
Whether traffic events there occurs with the pavement structure model automatic detection road surface for pre-building.Therefore, the friendship that the present invention is provided
Logical event detecting method is instead of in prior art to be observed monitor video image by staff on artificial judgment road surface to be
The no method that traffic events occur.This method by traffic incident detecting system automatic detection traffic events is saved in a large number
Human and material resources, reduce the working strength of monitoring personnel.And will not occur to cause to traffic due to deficient in energy
The wrong report of event and the risk that fails to report, therefore, the detection method improves the accuracy of traffic incidents detection.
Description of the drawings
In order to be expressly understood technical scheme, use during the specific embodiment that the present invention is described below
Accompanying drawing does a brief description.It should be evident that these accompanying drawings are only the section Example of the present invention, those of ordinary skill in the art
Other accompanying drawings can also be obtained on the premise of creative work is not paid.
Fig. 1 is the traffic incidents detection method flow schematic diagram that the embodiment of the present invention one is provided；
Fig. 2 is the method for building up schematic flow sheet of the pavement structure model that the embodiment of the present invention one is provided；
Fig. 3 is the traffic incidents detection method flow schematic diagram that the embodiment of the present invention two is provided；
Fig. 4 be the embodiment of the present invention two provide car week environmental model foundation method flow schematic diagram；
Fig. 5 is that the method flow using intelligent taking turn monitor mode detection traffic events that the embodiment of the present invention three is provided shows
It is intended to；
Fig. 6 is the operation stream that the intelligent taking turn monitoring system that the embodiment of the present invention three is provided is carried out under manual monitoring pattern
Journey schematic diagram；
Fig. 7 is the traffic incidents detection apparatus structure schematic diagram that the embodiment of the present invention four is provided；
Fig. 8 is that the structure of first specific embodiment of the traffic incidents detection device that the embodiment of the present invention four is provided is shown
It is intended to；
Fig. 9 is that the structure of second specific embodiment of the traffic incidents detection device that the embodiment of the present invention four is provided is shown
It is intended to；
Figure 10 is the knot of the traffic incidents detection device using intelligent taking turn monitoring system that the embodiment of the present invention four is provided
Structure schematic diagram.
Specific embodiment
For making the goal of the invention of the present invention, technical scheme and the technique effect that reaches clearer, complete, with reference to attached
Figure is described in detail to the specific embodiment of the present invention.
Embodiment one
Fig. 1 is the schematic flow sheet of the traffic incidents detection method that the embodiment of the present invention one is provided.As shown in figure 1, the friendship
Logical event detecting method is comprised the following steps：
The Traffic Surveillance Video image of S101, acquisition video camera in gathering when presetting presetting bit：
Video camera of the control on road is opened, and makes the traffic information on camera acquisition road.Due to video camera
It is typically mounted on head, so video camera can be according to demand in the default presetting bit that specifies.When video camera is opened
Afterwards, video camera can just gather traffic information, and the traffic information for collecting is uploaded to intelligent monitor and control center, in intelligent monitoring
The heart can get Traffic Surveillance Video image of the video camera in gathering when presetting presetting bit.
S102, track of vehicle model is set up according to the Traffic Surveillance Video image：
It should be noted that can be set up according to the Traffic Surveillance Video image using technological means customary in the art
Track of vehicle model.
As a specific embodiment of the present invention, described track of vehicle is set up according to the Traffic Surveillance Video image
Model specifically may comprise steps of：
S1021, the vehicle detected in the Traffic Surveillance Video image：
It should be noted that the method for vehicles in detection video image provided in an embodiment of the present invention can adopt this area
The method of usual employing background modeling.This using background modeling method by being modeled to the background of image, model is built
After vertical, present image be compared with background image, foreground target is determined according to comparative result.But the method exists following
Problem：
1) sensitive to shade, it is easy to be moving target by shade error detection；
2) adaptability under low contrast and halation environment is poor；
3) to object nothing separating capacity, misrecognition is also easy to produce, so as to have a strong impact on the statistics of the magnitude of traffic flow and speed
As a result problem.
In order to avoid the problems referred to above, (publication date is in September, 2007 present invention preferably employs Publication No. CN101030256A
5 days) Chinese patent application.The method of the patent application can definitely recognition detection target, compared to traditional based on the back of the body
The method of scape modeling, the method suffer from clear superiority at aspects such as shadow interference, complex illuminations.
S1022, the vehicle to detecting are tracked, and obtain the movement locus of the vehicle：
Vehicle to detecting is tracked, and its movement locus is analyzed, and can definitely understand the row of vehicle
Sail track.
The wireless vehicle tracking that the present invention is adopted can be adopted based on optical flow tracking or the method for template matches, but this
A little methods are usually present problems with：1) occlusion issue, 2) deformation problems, vehicle would generally occur size during traveling
Change, while turn that the metamorphosis of vehicle is also resulted in, so as to affect the stability for tracking.
In order to avoid the problems referred to above, present invention preferably employs a kind of new wireless vehicle tracking, the method using tracking and
The mechanism that detection combines, and the real-time model of target vehicle is set up by on-line study, while will be anti-for the result of on-line study
Act on tracking and detect, final target positioning is determined by tracking and detection collaboration.The method can with effectively solving vehicle with
The deformation of target during track, the blocking and block rear vehicle due to the problem that deforms upon such as turn of other objects.
S1023, the movement locus according to the vehicle, set up track of vehicle model.
S103, according to the track of vehicle model and the pavement structure model inspection traffic events for pre-building：
It should be noted that the mould that monitor video image of the pavement structure model according to road traffic situation when normal is set up
Type.
By vehicle movement locus of statistics road traffic situation when normal, track of vehicle is modeled, and is carried out
Line learns, and so as to automatically identify pavement structure, sets up corresponding pavement structure model according to the pavement structure.The road surface is handed over
The normal situation of logical situation refers to the situation that traffic events do not occur on road surface.
It should be noted that the movement locus of vehicle generally includes two factors：Travel direction and travel speed.So,
The pavement structure model includes the direction of motion angle model of vehicle and the movement velocity model of vehicle, the track mould of same vehicle
Type also includes the direction of motion angle model of vehicle and the movement velocity model of vehicle.Therefore, according to locus model and the road of vehicle
Face structural model can detect the traffic events relevant with the travel direction of vehicle and travel speed.
In embodiments of the present invention, described according to track of vehicle model and the pavement structure model inspection traffic that pre-builds
Event is specially：
Obtain the track of vehicle model and the pavement structure model：
Compare the track of vehicle model and the direction angle model in the pavement structure model and/or rate pattern, with
Obtain the direction angular displacement in the track of vehicle model and the pavement structure model and/or velocity deviation；
Judge whether the direction angular displacement reaches first pre-conditioned, and/or, judge whether the velocity deviation reaches
Second is pre-conditioned, to judge whether traffic events occur on road surface.
Inclined when the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model
When difference is more than angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is big
When First Speed threshold value, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is big
When second speed threshold value, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, sentence
Surely there is vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed
During threshold value, judge vehicle congestion event；
When the time that the track of vehicle model of at least two vehicles on road surface overlap occurs and overlaps is continued above second
Between section when, judge occur vehicle crash event.
It should be noted that in embodiments of the present invention, in order to be more accurately detected traffic events, traffic monitoring is regarded
Each two field picture in frequency image becomes multiple fritters according to identical regular partition, and pavement structure model is by the plurality of fritter
Pavement structure model constitutes.
Used as a specific embodiment of the present invention, each two field picture in Traffic Surveillance Video image is advised according to identical
N × M fritter is then divided into, and wherein, N, M are the integer more than or equal to 1；
The method for building up all same of the pavement structure model of each fritter in monitor video image, below with division
Pavement structure model is described as a example by the method for building up of the pavement structure model of any one fritter in N × M fritter
Method for building up.
Set and fritter is preset as any one fritter in N × M fritter, if it is considered to N × M fritter is in monitor video
If an array being constituted in image, then preset fritter position in the array and arrange for the i-th row jth, the default fritter is in prison
Position in control video image can be expressed as (i, j), wherein, 1≤i≤N, 1≤j≤M, and i, j are integer.
As a specific embodiment of the present invention, as shown in Fig. 2 the foundation of the pavement structure model of the default fritter
Method is comprised the following steps：
S21, extract from the t two field picture and t-1 two field picture of default fritter for track characteristic point：
Using technological means customary in the art, extracting from the t two field picture and t-1 two field picture of default fritter is used for
The characteristic point of tracking.
In embodiments of the present invention, t >=2, and t is integer.
S22, the movement locus to the characteristic point for tracking in t two field picture and t-1 two field picture carry out light stream and estimate
Meter, obtains the initialization model of the pavement structure in the default fritter.
S23, the initialization model is optimized using the t+1 two field picture in the default fritter, is preset
Pavement structure model after renewal in fritter.
S24, t is judged whether less than default training frame number, if not, execution step S25.If it is, execution step
S26.
S25, to described update after pavement structure model carry out model filtering, finally give the road surface of the default fritter
Structural model：
Pavement structure model after renewal is carried out one of purpose of model filtering be in order to eliminate the fortune of pedestrian on road
Dynamic rail mark and the track of vehicle not travelled according to traffic rules so that the pavement structure model for obtaining meets normal traffic conditions
Under information of road surface.
S26, t two field picture is updated to t+1 two field picture, returns and execute step S21.
It should be noted that step S21 to step S26 is the foundation of the pavement structure model in any one fritter
Journey, it is also possible to regard the training process of the pavement structure model in any one fritter as.
Statistics shows that the deflection θ of road vehicles motion meets Gaussian Profile with movement velocity v of vehicle, therefore, every
The direction of vehicle movement of each fritter and the mixing height of speed is set up in the training process of the pavement structure model of one fritter
This model P(i,j)(θ,v).
In embodiments of the present invention, the pavement structure model of vehicle can be regarded as the pavement structure model of fritter shared by which
Set.
In embodiments of the present invention, in order that the generation of intelligent monitor system automatic detection is various types of on road
Traffic events, are previously provided with various types of traffic events models in intelligent monitor system.When road traffic situation meets certain
After the traffic events model of one type, then intelligent monitor system judges to there occurs the traffic events of the type on road surface automatically.
As described above, in embodiments of the present invention, traffic events include that vehicle drives in the wrong direction event, overspeed of vehicle event, vehicle
Slow event, vehicle stopping event, vehicle congestion event and vehicle crash event.
Accordingly, each traffic events model above-mentioned is as follows：
Before each traffic events model is described, the physics meaning of the parameters under explaining first in following model
Justice.(i, j) is that position of the fritter in monitor video image is arranged for the i-th row jth；N represents the characteristic point in fritter for tracking
Number；M represents the shared little block number in monitor video image of vehicle.
Wherein, the retrograde event model of vehicle is：
Wherein,Represent average angle of the pavement structure model in the fritter (i, j)；
Represent angle intermediate value of the vehicle in the fritter；
δ is angle threshold.
Overspeed of vehicle event model is：
Wherein,Represent average speed of the pavement structure model in fritter (i, j)；
For speed intermediate value of the vehicle in track of vehicle model in the fritter (i, j)；
ε 1 represents First Speed threshold value.
Automobile low-speed event model is：
Wherein,Represent average speed of the pavement structure model in fritter (i, j)；
For speed intermediate value of the vehicle in track of vehicle model in the fritter (i, j)；
ε 2 represents second speed threshold value.
Vehicle stopping event model in first time period, speed variables in track of vehicle model and pavement structure model
In the relation of speed variables meet formula below：
Wherein,Represent average speed of the pavement structure model in fritter (i, j)；
For speed intermediate value of the vehicle in track of vehicle model in the fritter (i, j)；
ε 3 represents third speed threshold value, and ε 3 is almost nil.
Model is stopped according to vehicle and can be determined that out that vehicle occurs Parking, using same decision method, when
After when most of vehicles could be arranged to all occur to stop more than the vehicle of certain predetermined ratio on road, road is now judged
On there occurs traffic congestion event.
When overlapping occurs in the track of vehicle model of at least two vehicles on road, and at least two vehicles stop,
Then intelligent monitor system judges to there occurs vehicle crash event on road.
Further, after judging to there occurs vehicle crash event on road, can be further according to the car for colliding
Track of vehicle model in direction angle model and rate pattern judging vehicle collision type：Side collision, knock into the back or just
Face is collided.
S104, alarm is sent when traffic events are detected.
As the preferred embodiments of the present invention, when detect there occurs traffic events on road surface after, can also send out
Alarm, to point out traffic events of the staff to detecting to process.
The traffic incidents detection method provided by embodiment one, intelligent monitor system is by comparing locus model and road
Direction angle model and/or travel speed model in the structural model of face, then according in locus model and pavement structure model
It is first pre-conditioned whether direction angular displacement meets, and/or, whether travel speed deviation meets second pre-conditioned (i.e. road surface
Whether traffic reaches traffic events model) judging whether there occurs traffic events on road surface.The traffic that the present invention is provided
Event detecting method can be on automatic detection road surface traffic events.Therefore, the traffic incidents detection method that the present invention is provided is replaced
Whether there are traffic events for monitor video image being observed by staff in prior art on artificial judgment road surface
Method.This method by traffic incident detecting system automatic detection traffic events saves substantial amounts of man power and material money
Source, reduces the working strength of monitoring personnel.And will not occur to cause the wrong report to traffic events as personnel are deficient in energy
With the risk that fails to report, therefore, the detection method improves the accuracy of traffic incidents detection.
In traffic incidents detection method described in embodiment one, it is travel direction and the row of vehicle according to track of vehicle model
The comparison for sailing direction of traffic angle model and rate pattern in speed and pavement structure model can detect on road surface whether send out
The traffic events related to track of vehicle are given birth to, for example：Vehicle drive in the wrong direction, hypervelocity, at a slow speed, stop, collision etc. event.Except this it
Outward, present invention also offers a kind of can detect the method for shedding thing.Referring specifically to embodiment two.
Embodiment two
Fig. 3 is the schematic flow sheet of the traffic incidents detection method that the embodiment of the present invention two is provided, as shown in figure 3, the friendship
Logical event detecting method is comprised the following steps：
The Traffic Surveillance Video image of S301, acquisition video camera in gathering when presetting presetting bit：
The step is identical with step S101 in embodiment one, and for the sake of brevity, here is not described in detail, concrete ginseng
See the description of embodiment one.
S302, set up according to the Traffic Surveillance Video image car week environmental model：
As shown in figure 4, the step specifically includes following steps：
S41, vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image：
Vehicle and surrounding environment portion are chosen from the Traffic Surveillance Video image using technological means customary in the art
Point.
S42, according to the car week and surrounding environment part obtain car week environment contour feature figure：
Specifically, according to the vehicle that chooses and surrounding environment part, profile diagram is obtained based on sobel and SURF algorithm, then
Lane line interference is removed by Hough algorithm, obtains car week environment contour feature figure.
S43, according to the car week environment contour feature figure set up car week environment DPM model：
Car body is partitioned into from car week environment contour feature in figure, auto model p is set up according to car body0, from the car
All environment contour feature in figures are partitioned into the surrounding environment of vehicle, before and after setting up vehicle respectively according to the surrounding environment being partitioned into
Model on the four direction of left and right.
With auto model p0As root model, the surrounding environment model using vehicle as partial model, this five model group
Become the DPM model of car week environment, wherein, surrounding environment model is the partial model of DPM model, and which is respectively the first partial model
p1, the second partial model p2, the 3rd partial model p3With the 4th partial model p4.The DPM for being formulated car week environment (can make
Make) model is as follows：
Z=(p0,p1,p2,p3,p4).
S303, according to the car week environmental model and the pavement structure model inspection for pre-building shed formal matter part：
The step specifically includes following steps：
S3031, vehicle environmental model is mated with pavement structure model, obtained matching value：
It should be noted that the pavement structure model described in the embodiment of the present invention includes the root model being made up of auto model
And four partial models of vehicle's surroundings environment structure.
Wherein, the formula mated with pavement structure model by vehicle environmental model is as follows：
Wherein, score (p0,p1,p2,p3,p4) it is matching value；
F0, FiRepresent root model and the i-th partial model in pavement structure model respectively.
S3032, when matching value more than predetermined threshold value when, it is determined that car Zhou Yichang, from car week environment in be partitioned into Che Zhouyi
Normal object：
S3033, the movement locus to car Zhou Yichang object are tracked, and set up the locus model of car Zhou Yichang object：
S3034, when the medium velocity of the locus model of car Zhou Yichang object is 0, determine car Zhou Yichang object for resting
Body, and start timing, after the time of staying of car Zhou Yichang object Preset Time is reached, feature is carried out to car Zhou Yichang object and is carried
Take and classification：
In order to more accurately determine car Zhou Yichang object for shedding thing, when the time of staying of car Zhou Yichang object reach pre-
If after the time, car Zhou Yichang object is likely to be pedestrian, in order to avoid the erroneous judgement to shedding thing, is determining car Zhou Yichang object
Whether it is that preferred pair car Zhou Yichang object carries out feature extraction and classifying before shedding thing.
S3035, according to extract feature determine whether car Zhou Yichang object is to shed thing.
S304, alarm is sent when detecting and shedding formal matter part.
As a preferred embodiment of the present invention, when detect on road surface there occurs shed formal matter part after, can also be to
Alarm is sent outward, to remind staff.
By the traffic incidents detection method described in embodiment two can automatic detection go out on road surface shed formal matter part.
By the traffic incidents detection method described in embodiment one and embodiment two, can not only detect with car and connect track
Relevant traffic events, additionally it is possible to detected according to car week environment and shed formal matter part.The friendship that embodiment one and embodiment two are detected
Interpreter's part has consisted essentially of all traffic events on road, so the traffic thing provided by embodiment one and embodiment two
Part detection method, intelligent monitor system can automatic detection go out traffic events on road occur, therefore, embodiment one and real
The method that the traffic incidents detection method of the offer of example two instead of manual detection traffic events is applied, improves detection traffic events
Accuracy, reduces the risk for missing inspection or flase drop occur.
It should be noted that in embodiment one and the traffic incidents detection method described in embodiment two, video camera can one
A certain presetting bit is directly fixed on, and during whole monitoring work, camera supervised visual angle immobilizes, but this fixation
The monitor mode at visual angle, it is possible to there is monitoring blind area, so as to cause failing to report for traffic events.Blind in order to avoid there is monitoring
Area, and the feature of monopod video camera is made full use of, as a preferred embodiment of the present invention, can be using intelligent taking turn monitoring
Method control rotated between multiple different presetting angles per road video camera.Per the multiple differences residing for the video camera of road
The monitoring visual field that reaches of presetting angle can cover the cross section on whole road surface.This examined using intelligent taking turn monitor mode
The method of traffic events is surveyed as shown in embodiment three.
It should be noted that when multichannel video camera is provided with a certain section, the position control side of the multichannel video camera
Formula can be to adopt coordinated signals mode.When the position of a road video camera changes, the position of other road video cameras also occurs
Change.In order to avoid monitoring blind area, the multichannel video camera being arranged on same section can be protected in synchronization location
Card monitoring visual field covers the cross section on whole road surface.In addition, when multichannel video camera carries out rotation between different presetting bits, energy
Enough ensure the seamless connection of monitoring visual field.
Embodiment three
It should be noted that the mode of operation of the intelligent taking turn monitoring system described in the embodiment of the present invention includes automatic taking turn
Monitoring mode and manual monitoring pattern.Under normal circumstances, intelligent taking turn monitoring system is in automatic taking turn monitoring mode, only exists
In particular cases as there are traffic events, manual monitoring pattern when needing manual intervention, can be just in.
Fig. 5 be the embodiment of the present invention three provide detect the method flow of traffic events using intelligent taking turn monitor mode
Schematic diagram.As shown in figure 5, the method for detecting traffic events using intelligent taking turn monitor mode is comprised the following steps：
S501, the shooting angle of adjustment video camera reach a certain default presetting bit, and the default presetting bit is regarded as shooting
The current presetting bit of machine：
The shooting angle of intelligent monitor system adjustment video camera reaches a certain default presetting bit, and the default presetting bit is regarded as
It is the current presetting bit of video camera.
S502, the associated profile of the current presetting bit of loading.
S503, according to the associated profile of current presetting bit detect traffic events.
It should be noted that the associated profile detection traffic events of the current presetting bit of basis described in this step are concrete
Including step S101 described in embodiment one to step S103, and/or step S301 described in embodiment two is to step S303.
S504, when traffic events are detected, send out alarm.
S505, after the instruction that staff carries out manual intervention is received, intelligent taking turn monitoring system is from automatic taking turn
Monitoring mode switches to manual monitoring pattern：
After monitoring work personnel receive the alarm that intelligent monitor system sends, monitoring work personnel determine whether
Whether the traffic events of warning are rational traffic events, to decide whether to carry out manual intervention.When intelligent taking turn monitoring system
After receiving the instruction that staff carries out manual intervention, intelligent taking turn monitoring system switches to people from automatic taking turn monitoring mode
Work monitoring mode.
After intelligent taking turn monitoring system switches to manual monitoring pattern from automatic taking turn monitoring mode, as shown in fig. 6, intelligence
Energy taking turn monitoring system can execute following operation：
S61, cancellation load presetting bit associated profile.
After S62, default manual monitoring duration to be achieved, whether intelligent taking turn monitoring system prompting loads presetting bit correlation is joined
Put file.
If after S63 obtains the positive reply of monitoring work personnel, intelligent taking turn monitoring system is cut from manual monitoring pattern
Shift to automatic taking turn monitoring mode.
If after S64 obtains the negative reply of monitoring work personnel, when intelligent taking turn monitoring system extends manual monitoring
Long, and return execution step S62.
S506, after the instruction that staff's refusal carries out manual intervention is received, intelligent taking turn monitoring system judges to take the photograph
Whether camera reaches Preset Time in the monitoring period of current presetting bit, if it is, execution step S507.
Current presetting bit and is updated to the next presetting bit to next one presetting bit by S507, control camera switching,
Return execution step S502.
The intelligent taking turn monitoring method provided by embodiment three can either make video camera play the advantage of automatic taking turn, and
The working strength of monitoring personnel can be substantially reduced, is farthest reduced event and fails to report and improve reaction speed so that take the photograph per platform
The monitoring capacity of camera is greatly improved.
The specific embodiment party of embodiment one to the detection method that embodiment three is traffic events provided in an embodiment of the present invention
Formula, the detection method of the traffic events provided based on above-described embodiment, present invention also offers the detection means of traffic events
Specific embodiment, referring specifically to following examples four.
Example IV
Fig. 7 is the structural representation of the traffic incidents detection device that the embodiment of the present invention four is provided.As shown in fig. 7, the friendship
Logical event detection device is included with lower unit：
Acquiring unit 71, for obtaining the Traffic Surveillance Video image gathered when video camera is in and presets presetting bit；
Unit 72 set up by model, for setting up track of vehicle model and/or car week according to the Traffic Surveillance Video image
Environmental model；
Detector unit 73, for according to the track of vehicle model and the pavement structure model inspection traffic thing for pre-building
Part, and/or, according to car week environmental model and the pavement structure model inspection traffic events for pre-building；The road surface knot
Structure model is the model that basis does not occur that monitor video image during traffic events is set up；
Traffic events be there occurs in order to remind on staff road surface, the traffic incidents detection device can also include：
Alarm unit 74, when traffic events are detected, sends alarm.
The traffic events on road surface can be automatically detected that by the traffic incidents detection device shown in Fig. 7, it is to avoid
Method by manually being detected, saves substantial amounts of human and material resources resource, reduces the working strength of monitoring personnel, and
And will not occur due to the deficient in energy risk for causing the wrong report to traffic events and failing to report, therefore, the detection means improves
The accuracy of traffic incidents detection.
In order to detect the traffic events related to track of vehicle, as first specific embodiment of the present invention, such as
Shown in Fig. 8, unit 72 set up by the model in traffic incidents detection device shown in above-described embodiment includes：
Detection sub-unit 721, for detecting the vehicle in the Traffic Surveillance Video image；
Tracking subelement 722, for being tracked to the vehicle for detecting, obtains the movement locus of the vehicle；
Subelement 723 set up by track of vehicle model, for the movement locus according to the vehicle, sets up track of vehicle mould
Type.
Further, the improvement further to above-mentioned first specific embodiment, as shown in figure 8, the track of vehicle mould
Type and the pavement structure model all include the direction angle model of vehicle motion and Velicle motion velocity model；
The detector unit 73 is specifically included：
Comparing subunit 731, for the direction angle mould in relatively the track of vehicle model and the pavement structure model
Type and/or rate pattern, direction angular displacement and/or the speed for obtaining the track of vehicle model and the pavement structure model are inclined
Difference；
Judgment sub-unit 732, first pre-conditioned for judging whether the direction angular displacement reaches, and/or, the speed
It is second pre-conditioned whether degree deviation reaches, to judge whether traffic events occur on road surface；
Inclined when the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model
When difference is more than angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is big
When First Speed threshold value, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is big
When second speed threshold value, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, sentence
Surely there is vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed
During threshold value, judge vehicle congestion event；
When the time that the track of vehicle model of at least two vehicles on road surface overlap occurs and overlaps is continued above second
Between section when, judge occur vehicle crash event.
By above-mentioned first specific embodiment, traffic incidents detection device is capable of detecting when the friendship related to track of vehicle
Interpreter's part, these traffic events include vehicle drive in the wrong direction, hypervelocity, at a slow speed, stopping, congestion, collision accident.
The traffic incidents detection device that the present invention is provided is gone back in addition to it can detect the traffic events related to track of vehicle
Can detect and shed formal matter part.
Formal matter part is shed in order to detect, as second specific embodiment of the present invention, as shown in figure 9, above-mentioned Fig. 8
Unit 72 set up by the model in shown traffic incidents detection device can also include：
Subelement 721 ' is chosen, for vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image；
Subelement 722 ' is obtained, for car week environment contour feature figure being obtained according to the vehicle and surrounding environment part；
Subelement 723 ' set up by all environment DPM models of car, for setting up car week according to car week environment contour feature figure
Environment DPM model, the car week environment DPM model include root model p0, the first partial model p1, the second partial model p2, the 3rd
Partial model p3With the 4th partial model p4, described model p0For auto model, the first partial model p1, second local
Model p2, the 3rd partial model p3With the 4th partial model p4Environmental model respectively on the four direction all around of vehicle.
Further, the detector unit 73 can also include：
Coupling subelement 731 ', for being mated vehicle environmental model with the pavement structure model, obtains coupling
Value；
Segmentation subelement 732 ', for when the matching value is more than predetermined threshold value, being partitioned into car week from all environment of car
Abnormal object；
Car Zhou Yichang object locus model sets up subelement 733 ', for entering to the movement locus of the car Zhou Yichang object
Line trace, sets up the locus model of car Zhou Yichang object；
Judge shed thing subelement 734 ', for the locus model when car Zhou Yichang object medium velocity be 0 when, start meter
When, after the time of staying of car Zhou Yichang object Preset Time is reached, determine the car Zhou Yichang object for shedding thing.
It should be noted that the traffic incidents detection device shown in Fig. 9 is in the traffic incidents detection device shown in Fig. 8
On the basis of the improvement that carries out, in fact, as another embodiment of the present invention, shedding the traffic events inspection of formal matter part for detection
Survey device to improve on the basis of the traffic incidents detection device shown in Fig. 7.Information according to disclosed in Fig. 9, this
Field those of ordinary skill's on the premise of not paying creative work can obtain the embodiment, and here is no longer retouched in detail
State.
In above-mentioned first specific embodiment and second specific embodiment, video camera can be permanently affixed to a certain preset
Position, during whole monitoring work, camera supervised visual angle immobilizes, but the monitor mode of this fixed viewpoint,
Monitoring blind area is there may exist, so as to cause failing to report for traffic events.In order to avoid there is monitoring blind area, and make full use of head
The feature of video camera, as a preferred embodiment of the present invention, can be taken the photograph per road using the method control of intelligent taking turn monitoring
Camera is rotated between multiple different presetting angles.Reach per the multiple different presetting angle residing for the video camera of road
Monitoring visual field can cover the cross section on whole road surface.This device of traffic events is detected using intelligent taking turn monitor mode
As shown in the 3rd specific embodiment.
It should be noted that when multichannel video camera is provided with a certain section, the position control side of the multichannel video camera
Formula can be to adopt coordinated signals mode.When the position of a road video camera changes, the position of other road video cameras also occurs
Change.In order to avoid monitoring blind area, the multichannel video camera being arranged on same section can be protected in synchronization location
Card monitoring visual field covers the cross section on whole road surface.In addition, when multichannel video camera carries out rotation between different presetting bits, energy
Enough ensure the seamless connection of monitoring visual field.
Further, the traffic incidents detection device is carried out certainly between the presetting bit of multiple differences using every road video camera
Driving wheel patrols the method for monitoring and carries out the detection of traffic events, can cover entirely per the multiple differences presetting bit residing for the video camera of road
The cross section on road surface, sets video camera after the monitoring period of a presetting bit reaches Preset Time, and automatic alternate is to the next one
Presetting bit is monitored.
Used as the 3rd specific embodiment of the present invention, as shown in Figure 10, described device is concrete except having first
Beyond the unit of embodiment or second specific embodiment, can also include with lower unit：
Cancel and presetting bit associated profile unit 101 is loaded, for manual intervention being carried out receiving staff
After instruction, cancel and load presetting bit associated profile；
Tip element 102, for reaching default artificial supervise from receiving staff and carrying out the instruction of manual intervention
After control duration, point out presetting bit associated profile whether is loaded, if it is, control camera switching is to the default presetting bit
Next presetting bit position, with the traffic shape on the corresponding road of the next presetting bit that gathers the default presetting bit
Condition, if not, extend artificial monitor duration, receive certainly extends artificial monitor duration instruction play again reach default manually
After monitor duration, the signal of manual monitoring duration is sent to the Tip element.
Further, described device can further include：Judging unit 103, refuses for receiving staff
After carrying out absolutely the instruction of manual intervention, judge whether video camera reaches Preset Time in the monitor event of current presetting bit, if
It is to control camera switching to the next presetting bit of the default presetting bit, and video camera location signal is sent
To the acquiring unit.
It should be noted that the traffic incidents detection device shown in Figure 10 be the present invention first specific embodiment i.e.
The improvement carried out on the basis of embodiment shown in Fig. 8.In fact, it is possible to the traffic incidents detection device for carrying out intelligent taking turn can
To improve on the basis of any of the above-described embodiment.
It is the preferred embodiments of the present invention above.It should be pointed out that for the person of ordinary skill of the art, do not taking off
On the premise of the inventive concept of the present invention, some improvements and modifications can also be made, these improvements and modifications also should be at this
The row of bright protection domain.
Claims (10)
1. a kind of detection method of traffic events, it is characterised in that the traffic incidents detection method is using existing per road video camera
The method for carrying out automatic taking turn monitoring between the presetting bit of multiple differences, can cover per the multiple differences presetting bit residing for the video camera of road
The cross section on whole road surface being covered, video camera being set after the monitoring period of a presetting bit reaches Preset Time, automatic alternate is arrived
Next presetting bit is monitored, and the traffic incidents detection method includes：
Obtain Traffic Surveillance Video image of the video camera in gathering when presetting presetting bit；
Track of vehicle model and/or car week environmental model are set up according to the Traffic Surveillance Video image；
According to the track of vehicle model and the pavement structure model inspection traffic events for pre-building, and/or, according to the car
All environmental models and the pavement structure model inspection traffic events for pre-building；The pavement structure model be according on road surface not
The model for occurring the monitor video image of traffic events to set up；
After the instruction for carrying out manual intervention is received, cancel and presetting bit associated profile is loaded, receiving certainly is carried out manually
The instruction of intervention is risen to reaching after default manual monitoring duration, points out whether load presetting bit associated profile, if it is, control
Camera switching processed to the position of the next presetting bit of the default presetting bit, to gather the next one of the default presetting bit
Traffic on the corresponding road of presetting bit, if not, extending artificial monitor duration, receives the artificial monitor duration of prolongation certainly
Instruction rise to being reached after default manual monitoring duration again, return and execute whether the prompting loads presetting bit relevant configuration text
The step of part；
After the instruction that refusal carries out manual intervention is received, judge whether video camera is reached in the monitor event of current presetting bit
Preset Time, if it is, control camera switching to the default presetting bit next presetting bit, and return execute described in obtain
The step of taking Traffic Surveillance Video image of the video camera in gathering when presetting presetting bit.
2. method according to claim 1, it is characterised in that described vehicle is set up according to the Traffic Surveillance Video image
Locus model includes：
Detect the vehicle in the Traffic Surveillance Video image；
Vehicle to detecting is tracked, and obtains the movement locus of the vehicle；
According to the movement locus of the vehicle, track of vehicle model is set up.
3. method according to claim 1, it is characterised in that the track of vehicle model and the pavement structure model equal
Direction angle model and Velicle motion velocity model including vehicle motion；
Described specifically included according to track of vehicle model and the pavement structure model inspection traffic events that pre-build：
Relatively the direction angle model in the track of vehicle model and the pavement structure model and/or rate pattern, obtain institute
State direction angular displacement and/or the velocity deviation of track of vehicle model and the pavement structure model；
Judge whether the direction angular displacement reaches first pre-conditioned, and/or, judge whether the velocity deviation reaches second
Pre-conditioned, to judge whether traffic events occur on road surface；
When the deviation of the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model is big
When angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is more than the
During one threshold speed, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is more than the
During two threshold speeds, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, judge to send out
Raw vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed threshold value
When, judge vehicle congestion event；
When the time that overlapping and overlapping occur in the track of vehicle model of at least two vehicles on road surface is continued above second time period
When, judge vehicle crash event.
4. method according to claim 1, it is characterised in that car week environment is set up according to the Traffic Surveillance Video image
Model is specifically included：
Vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image；
Car week environment contour feature figure is obtained according to the vehicle and surrounding environment part；
Car week environment DPM model is set up according to car week environment contour feature figure, the car week environment DPM model includes root mould
Type p0, the first partial model p1, the second partial model p2, the 3rd partial model p3With the 4th partial model p4, described model p0
For auto model, the first partial model p1, the second partial model p2, the 3rd partial model p3With the 4th partial model p4Point
Environmental model that Wei be on the four direction all around of vehicle.
5. method according to claim 4, it is characterised in that described according to car week environmental model and pre-build
Pavement structure model inspection traffic events are specifically included：
Vehicle environmental model is mated with the pavement structure model, is obtained matching value；
When matching value is more than predetermined threshold value, from all environment of car, it is partitioned into car Zhou Yichang object；
Movement locus to the car Zhou Yichang object is tracked, and sets up the locus model of car Zhou Yichang object；
When the medium velocity of the locus model of car Zhou Yichang object is 0, starts timing, reach when the time of staying of car Zhou Yichang object
To after Preset Time, determine the car Zhou Yichang object for shedding thing.
6. a kind of detection means of traffic events, it is characterised in that the traffic incidents detection device is using existing per road video camera
The method for carrying out automatic taking turn monitoring between the presetting bit of multiple differences carries out the detection of traffic events, per many residing for the video camera of road
The presetting bit of individual difference can cover the cross section on whole road surface, set video camera a presetting bit monitoring period reach default
After time, automatic alternate is monitored to next presetting bit, and the detection means includes：
Acquiring unit, for obtaining the Traffic Surveillance Video image gathered when video camera is in and presets presetting bit；
Unit set up by model, for setting up track of vehicle model and/or car week environment mould according to the Traffic Surveillance Video image
Type；
Detector unit, for according to the track of vehicle model and the pavement structure model inspection traffic events that pre-build, and/
Or, according to car week environmental model and the pavement structure model inspection traffic events for pre-building；The pavement structure model
It is the model that is set up according to monitor video image when there are no traffic events；
Cancel unit, after the instruction for carrying out manual intervention is received, cancel and load presetting bit associated profile；
Tip element, for reaching after default manual monitoring duration from the instruction for carrying out manual intervention is received, prompting is
No loading presetting bit associated profile；
Switch unit, for when presetting bit associated profile is loaded, control camera switching is to the default presetting bit
The position of next presetting bit, with the traffic on the corresponding road of the next presetting bit that gathers the default presetting bit；
Delay unit, for when presetting bit associated profile is not loaded with, extending artificial monitor duration, receiving certainly prolonged human
The instruction of work monitor duration is risen to being reached after default manual monitoring duration again, triggers the Tip element；
Judging unit, for, after the instruction that refusal carries out manual intervention is received, judging prison of the video camera in current presetting bit
Whether control event reaches Preset Time, if it is, next presetting bit of the control camera switching to the default presetting bit, and
Trigger the acquiring unit.
7. device according to claim 6, it is characterised in that unit set up by the model includes：
Detection sub-unit, for detecting the vehicle in the Traffic Surveillance Video image；
Tracking subelement, for being tracked to the vehicle for detecting, obtains the movement locus of the vehicle；
Subelement set up by track of vehicle model, for the movement locus according to the vehicle, sets up track of vehicle model.
8. device according to claim 6, it is characterised in that the track of vehicle model and the pavement structure model equal
Direction angle model and Velicle motion velocity model including vehicle motion；
The detector unit includes：
Comparing subunit, for the relatively track of vehicle model and the direction angle model in the pavement structure model and/or
Rate pattern, obtains direction angular displacement and/or the velocity deviation of the track of vehicle model and the pavement structure model；
Judgment sub-unit, first pre-conditioned for judging whether the direction angular displacement reaches, and/or, the velocity deviation
Whether reach second pre-conditioned, to judge whether traffic events occur on road surface；
When the deviation of the direction angle model in the direction angle model in the track of vehicle model with the pavement structure model is big
When angle threshold, judge the retrograde event of vehicle；
When the deviation of the speed variables in the speed variables in the track of vehicle model with the pavement structure model is more than the
During one threshold speed, judge overspeed of vehicle event；
When the deviation of the speed variables in the speed variables in the pavement structure model with the track of vehicle model is more than the
During two threshold speeds, judge automobile low-speed event；
When the speed variables in the track of vehicle model in first time period are continuously less than third speed threshold value, judge to send out
Raw vehicle stopping event；
When the speed variables in the track of vehicle model on road surface more than a certain proportion of vehicle are continuously less than third speed threshold value
When, judge vehicle congestion event；
When the time that overlapping and overlapping occur in the track of vehicle model of at least two vehicles on road surface is continued above second time period
When, judge vehicle crash event.
9. device according to claim 6, it is characterised in that unit set up by the model includes：
Subelement is chosen, for vehicle and surrounding environment part is chosen from the Traffic Surveillance Video image；
Subelement is obtained, for car week environment contour feature figure being obtained according to the vehicle and surrounding environment part；
Subelement set up by all environment DPM models of car, for setting up car week environment DPM mould according to car week environment contour feature figure
Type, the car week environment DPM model include root model p0, the first partial model p1, the second partial model p2, the 3rd partial model
p3With the 4th partial model p4, described model p0For auto model, the first partial model p1, the second partial model p2,
Three partial model p3With the 4th partial model p4Environmental model respectively on the four direction all around of vehicle.
10. device according to claim 6, it is characterised in that the detector unit includes：
Coupling subelement, for being mated vehicle environmental model with the pavement structure model, obtains matching value；
Segmentation subelement, for when the matching value is more than predetermined threshold value, being partitioned into car Zhou Yichang object from all environment of car；
Car Zhou Yichang object locus model sets up subelement, for being tracked to the movement locus of the car Zhou Yichang object,
Set up the locus model of car Zhou Yichang object；
Judge shed thing subelement, for the locus model when car Zhou Yichang object medium velocity be 0 when, start timing, work as car
After the time of staying of Zhou Yichang object reaches Preset Time, determine the car Zhou Yichang object for shedding thing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510182223.9A CN104809874B (en) | 2015-04-15 | 2015-04-15 | A kind of traffic incidents detection method and apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510182223.9A CN104809874B (en) | 2015-04-15 | 2015-04-15 | A kind of traffic incidents detection method and apparatus |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104809874A CN104809874A (en) | 2015-07-29 |
| CN104809874B true CN104809874B (en) | 2017-03-08 |
Family
ID=53694670
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510182223.9A Active CN104809874B (en) | 2015-04-15 | 2015-04-15 | A kind of traffic incidents detection method and apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104809874B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE102015217792B4 (en) * | 2015-09-17 | 2022-07-14 | Volkswagen Aktiengesellschaft | Device, method and computer program for providing information about the end of a traffic jam via a vehicle-to-vehicle interface |
| CN105206055B (en) * | 2015-09-24 | 2018-09-21 | 深圳市哈工大交通电子技术有限公司 | A kind of accident detection method of Traffic Surveillance Video identification vehicle collision |
| CN105355039A (en) * | 2015-10-23 | 2016-02-24 | 张力 | Road condition information processing method and equipment |
| CN105761547A (en) * | 2016-03-28 | 2016-07-13 | 安徽云森物联网科技有限公司 | Traffic collision pre-warning technique and system based on images |
| CN106101641B (en) * | 2016-07-18 | 2022-06-14 | 中国科学院地理科学与资源研究所 | Video monitoring system and monitoring method thereof |
| CN108307315B (en) * | 2016-09-07 | 2020-08-25 | 北京嘀嘀无限科技发展有限公司 | Traffic accident processing method, server and mobile terminal |
| CN106340205A (en) * | 2016-09-30 | 2017-01-18 | 广东中星微电子有限公司 | Traffic monitoring method and traffic monitoring apparatus |
| CN106448162B (en) * | 2016-09-30 | 2020-07-17 | 中星技术股份有限公司 | Road monitoring method and road monitoring device |
| CN107010086B (en) * | 2017-03-29 | 2018-07-10 | 宁夏凯速德科技有限公司 | High ferro line security control method and system |
| CN107274699A (en) * | 2017-06-30 | 2017-10-20 | 重庆交通大学 | Based on what position was shared passing method is cooperateed with without signal-control crossing vehicle certainly |
| CN107705595B (en) * | 2017-09-25 | 2020-11-27 | 航天科技控股集团股份有限公司 | Automatic control vehicle-mounted terminal triggering alarm device and triggering alarm method based on recorder management platform |
| CN107742425B (en) * | 2017-09-29 | 2021-02-12 | 公安部交通管理科学研究所 | Method for rapidly discovering abnormal road conditions based on road monitoring video |
| CN109934075A (en) * | 2017-12-19 | 2019-06-25 | 杭州海康威视数字技术股份有限公司 | Accident detection method, apparatus, system and electronic equipment |
| CN109035776A (en) * | 2018-08-23 | 2018-12-18 | 深圳大学 | A kind of accident detection method and system |
| CN111402612A (en) * | 2019-01-03 | 2020-07-10 | 北京嘀嘀无限科技发展有限公司 | Traffic incident notification method and device |
| CN110113579B (en) * | 2019-05-30 | 2021-04-16 | 浙江大华技术股份有限公司 | Method and device for tracking target object |
| CN111145378B (en) * | 2019-12-30 | 2021-06-25 | 华设设计集团股份有限公司 | Automatic video inspection system and method for road traffic incident |
| CN111753634A (en) * | 2020-03-30 | 2020-10-09 | 上海高德威智能交通系统有限公司 | Traffic incident detection method and device |
| CN111640309A (en) * | 2020-05-22 | 2020-09-08 | 浙江工贸职业技术学院 | Swift vehicle detecting system |
| CN112435470A (en) * | 2020-11-11 | 2021-03-02 | 宁波职业技术学院 | Traffic incident video detection system |
| CN112367475B (en) * | 2021-01-15 | 2021-03-30 | 上海闪马智能科技有限公司 | Traffic incident detection method and system and electronic equipment |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9183744B2 (en) * | 2008-01-29 | 2015-11-10 | Here Global B.V. | Method for providing images of traffic incidents |
| JP2013161281A (en) * | 2012-02-06 | 2013-08-19 | Sumitomo Electric Ind Ltd | Traffic image acquisition device, traffic image acquisition method, and traffic image acquisition program |
| CN102945603B (en) * | 2012-10-26 | 2015-06-03 | 青岛海信网络科技股份有限公司 | Method for detecting traffic event and electronic police device |
| CN103366569B (en) * | 2013-06-26 | 2015-10-07 | 东南大学 | The method and system of real-time grasp shoot traffic violation vehicle |
| CN103971521B (en) * | 2014-05-19 | 2016-06-29 | 清华大学 | Road traffic anomalous event real-time detection method and device |
-
2015
- 2015-04-15 CN CN201510182223.9A patent/CN104809874B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN104809874A (en) | 2015-07-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104809874B (en) | A kind of traffic incidents detection method and apparatus | |
| CN105799740B (en) | A kind of track foreign body intrusion automatic detection and method for early warning based on technology of Internet of things | |
| CN104691559B (en) | Monitoring station platform safety door and car door interval safe accessory system and its implementation | |
| CN101458871B (en) | Intelligent traffic analysis system and application system thereof | |
| CN104847211B (en) | Auxiliary system for safety of platform safety doors and train door sections and implementation method thereof | |
| CN103770780B (en) | A kind of active safety systems of vehicles alarm shield device | |
| CN101465033B (en) | Automatic tracking recognition system and method | |
| CN103366571B (en) | The traffic incidents detection method at night of intelligence | |
| CN106652551A (en) | Parking stall detection method and device | |
| CN102945603B (en) | Method for detecting traffic event and electronic police device | |
| CN105989682A (en) | Safety early warning and monitoring system and monitoring method for construction machinery under power transmission line | |
| KR102105162B1 (en) | A smart overspeeding vehicle oversee apparatus for analyzing vehicle speed, vehicle location and traffic volume using radar, for detecting vehicles that violate the rules, and for storing information on them as videos and images, a smart traffic signal violation vehicle oversee apparatus for the same, and a smart city solution apparatus for the same | |
| CN107134147A (en) | Method, apparatus and system based on video camera and sensor management bar parking space | |
| CN104809887B (en) | A kind of retrograde detection method of vehicle on expressway and autoalarm | |
| CN106485927A (en) | A kind of intelligent transportation violation information harvester and acquisition method | |
| CN109598187A (en) | Obstacle recognition method, differentiating obstacle and railcar servomechanism | |
| CN105632186A (en) | Method and device for detecting vehicle queue jumping behavior | |
| EP3112900B1 (en) | Safety system and method to detect a risk condition in a region to be monitored placed close to an operating machine, such as a drilling machine or the like | |
| CN106327863A (en) | Method, device and system for detection and early warning of vehicle congestion | |
| CN107054223A (en) | It is a kind of to be shown based on the driving blind area that panorama is shown and caution system and method | |
| CN103366569A (en) | Method and system for snapshotting traffic violation vehicle in real time | |
| CN110466562A (en) | There are detection devices and method for a kind of level crossing based on infrared laser and microwave | |
| CN111225189A (en) | Middle and small-sized channel bridge monitoring device | |
| CN112367475B (en) | Traffic incident detection method and system and electronic equipment | |
| CN115603466A (en) | Ship shore power system based on artificial intelligence visual recognition and control method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20171120 Address after: 518057 the South District of Nanshan District high tech Industrial Park, Shenzhen City, Guangdong Province, R1-B Patentee after: Shenzhen Neusoft Software Co., Ltd. Address before: Hunnan rookie street Shenyang city Liaoning province 110179 No. 2 Patentee before: Neusoft Group Corporation |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20181213 Address after: 110179 2 Hsin street, Hunnan New District, Shenyang, Liaoning Patentee after: Neusoft Group Corporation Address before: 518057 Block R1-B, South District of Nanshan High-tech Industrial Park, Shenzhen City, Guangdong Province Patentee before: Shenzhen Neusoft Software Co., Ltd. |