CN108717521A - A kind of parking lot order management method and system based on image - Google Patents
A kind of parking lot order management method and system based on image Download PDFInfo
- Publication number
- CN108717521A CN108717521A CN201810341890.0A CN201810341890A CN108717521A CN 108717521 A CN108717521 A CN 108717521A CN 201810341890 A CN201810341890 A CN 201810341890A CN 108717521 A CN108717521 A CN 108717521A
- Authority
- CN
- China
- Prior art keywords
- image
- parking lot
- abnormal object
- event
- vehicle
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B19/00—Alarms responsive to two or more different undesired or abnormal conditions, e.g. burglary and fire, abnormal temperature and abnormal rate of flow
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Abstract
The present invention discloses a kind of parking lot order management method and system based on image, is related to intelligent parking lot management domain, includes the following steps:Acquire each area monitoring image in parking lot;Analysis is identified to each area monitoring image, obtains the corresponding goal behavior information of each area monitoring image;Judged to whether there is abnormal object image in each area monitoring image according to the corresponding goal behavior information of each area monitoring image;When judgement is there are when abnormal object image, the abnormal object image is highlighted, and carry out abnormality processing.The method and system provided through the invention can promote parking lot management efficiency and abnormal judging nicety rate, reduce cost of labor, realize parking lot order automatic management.Method and system provided by the invention can be adapted for parking lot order management domain.
Description
Technical field
The present invention relates to intelligent parking lot management domain, more particularly to a kind of parking lot order management method based on image
And system.
Background technology
With being on the increase for automobile quantity, the efficient order management in parking lot has become the difficulty of parking management industry
Topic.Due to efficient management means deficiency, the behavior of various parkings lack of standardization and interference normal order is frequently present of inside and outside parking lot,
Driving channel, vehicle are such as occupied in entrance delay, crowd massing, object blocking.
Currently, there are two types of parking lot order way to manage is usual, one is traditional way to manages, by being set in entrance
It sets sentry box and personnel is arranged to be managed and dredge, arrange inspection administrator manually to be gone on patrol in parking lot, note abnormalities
Directly carry out artificial treatment and advice etc..Another mode is to utilize monitoring camera Added Management, by each in parking lot
Monitoring camera is installed in region, is supervised to the multi-channel video of monitoring center in field by way of artificial poll.
However the first above-mentioned way to manage causes human cost height, pipe due to needing a large amount of personnel and manual operation
Manage inefficiency.And second of way to manage is due to needing by way of artificial poll from the collected big spirogram of monitoring camera
Abnormal object image is identified as in, and abnormality processing, while the monitoring camera laid are carried out by artificial mode to scene
Generally only provide single mobile detection and detection function, play the role of it is identical as common safety monitoring, therefore use the pipe
During reason mode still needs more personnel to participate in abnormal judgement and manage on the spot, lead to the efficiency of management and exception of the pattern
Judging nicety rate is relatively low, and cost of labor is higher.
Invention content
In order to solve the above technical problems, the present invention provides a kind of parking lot order management method and system based on image,
It can solve the existing parking lot order efficiency of management and abnormal judging nicety rate is relatively low, problem with high labor costs.
To achieve the above object, on the one hand, the present invention provides a kind of parking lot order management method based on image, described
Method includes:
Acquire each area monitoring image in parking lot;
Analysis is identified to each area monitoring image, obtains the corresponding goal behavior of each area monitoring image
Information;
According to the corresponding goal behavior information of each area monitoring image judge in each area monitoring image whether
There are abnormal object images;
When judgement is there are when abnormal object image, the abnormal object image is highlighted and carries out exception
Reason.
Further, described that analysis is identified to each area monitoring image, obtain each area monitoring image difference
The step of corresponding goal behavior information includes:
Identify the corresponding target image classification of each area monitoring image;
By the corresponding target identification parser of different target image category, to each area monitoring image into
Row discriminance analysis obtains the corresponding goal behavior information of each area monitoring image.
Further, the target image classification includes vehicle image, and/or pedestrian image, and/or parking lot environment map
Picture;The target identification parser includes algorithm of target detection, target tracking algorism, goal behavior parser;The mesh
It includes vehicle detecting algorithm, and/or pedestrian detection algorithm, and/or environment measuring algorithm to mark detection algorithm;The goal behavior point
It includes vehicle behavior parser, and/or pedestrian behavior parser, and/or ambient condition parser to analyse algorithm.
Further, by the corresponding target identification parser of different target image category, to each region
Monitoring image is identified the step of analysis and includes:
If the target image classification is vehicle image, pass through vehicle detecting algorithm, target tracking algorism, vehicle behavior
Analysis is identified to the vehicle image in parser.
Further, by the corresponding target identification parser of different target image category, to each region
Monitoring image is identified the step of analysis and includes:
If the target image classification is pedestrian image, pass through pedestrian detection algorithm, target tracking algorism, pedestrian behavior
Analysis is identified to the pedestrian image in parser.
Further, by the corresponding target identification parser of different target image category, to each region
Monitoring image is identified the step of analysis and includes:
If the target image classification is parking lot ambient image, pass through environment measuring algorithm, target tracking algorism, environment
Analysis is identified to the parking lot ambient image in state analysis algorithm.
Further, each area monitoring figure is judged according to the corresponding goal behavior information of each area monitoring image
Include with the presence or absence of the step of abnormal object image as in:
According to the corresponding goal behavior information of synchronization associated region monitoring image, each area monitoring is judged
It whether there is abnormal object image in image.
Further, each area monitoring figure is judged according to the corresponding goal behavior information of each area monitoring image
Include with the presence or absence of the step of abnormal object image as in:
According to the corresponding goal behavior information of the same area monitoring image in preset time period, each region is judged
It whether there is abnormal object image in monitoring image.
Further, before the step of progress abnormality processing, the method further includes:
The corresponding event type of abnormal object image is determined according to the corresponding goal behavior information of abnormal object image.
Further, the step of progress abnormality processing includes:
Corresponding abnormality processing is carried out according to the corresponding event type of the abnormal object image.
Further, the abnormality processing strategy includes:
In the corresponding regional location outputting alarm information of the abnormal object image;And/or
Prompt message is solved in the corresponding regional location output abnormality of the abnormal object image;And/or
Control instruction is sent to the equipment of the corresponding regional location of the abnormal object image.
Further, the event type include vehicle traveling anomalous event, and/or pedestrian behavior anomalous event, and/or
Vehicle flowrate anomalous event, and/or parking lot environmental abnormality event.
Further, described the step of corresponding abnormality processing is carried out according to the abnormal object image corresponding event type
Including:
Anomalous event is travelled if the corresponding event type of the abnormal object image is vehicle, and/or pedestrian behavior is abnormal
Event, and/or parking lot environmental abnormality event, then in the corresponding regional location outputting alarm information of the abnormal object image.
Further, described the step of corresponding abnormality processing is carried out according to the abnormal object image corresponding event type
Including:
If the corresponding event type of the abnormal object image is vehicle flowrate anomalous event and/or parking lot environmental abnormality thing
Part then solves prompt message in the corresponding regional location output abnormality of the abnormal object image.
Further, described the step of corresponding abnormality processing is carried out according to the abnormal object image corresponding event type
Including:
If the corresponding event type of the abnormal object image is parking lot environmental abnormality event and/or vehicle flowrate exception thing
Part sends control instruction to the equipment of the corresponding regional location of the abnormal object image.
Further, if the event type, which is vehicle, travels anomalous event, the method further includes:
Identify the corresponding vehicle identification information of the abnormal object image;
By the vehicle license plate information preservation in abnormal vehicle information database.
Further, if the event type is pedestrian behavior anomalous event, the method further includes:
Identify car owner's characteristic information in the abnormal object image;
Car owner's characteristic information is stored in abnormal owner information database.
On the other hand, the present invention provides a kind of parking lot order management system based on image, the system comprises:Image
Harvester, image storage module, image analysis module, event analysis module, decision configuration module, abnormality processing execution are set
Standby, network communication module;
Image collecting device, for acquiring each area monitoring image in parking lot;
Image storage module, each area monitoring image in parking lot for storing image acquisition device;
Image analysis module, for image acquisition device or image storage module storage each region
Analysis is identified in monitoring image, obtains the corresponding goal behavior information of each area monitoring image;
Event analysis module, for judging each area according to the corresponding goal behavior information of each area monitoring image
It whether there is abnormal object image in the monitoring image of domain;
Decision configuration module, for arrangement abnormalities processing strategy;
Abnormality processing executes equipment, for when judgement is there are when abnormal object image, being carried out to the abnormal object image
It highlights, and abnormality processing is carried out according to the abnormality processing strategy of decision configuration module configuration;
Network communication module, for supporting image collecting device, image storage module, image analysis module, event analysis
Module, decision configuration module, abnormality processing execute and carry out information exchange between equipment, network communication module.
Further, described image analysis module includes:Front-end image analytic unit, image sorting unit, back-end image
Analytic unit;
Front-end image analytic unit, for identification corresponding target image classification of each area monitoring image;
Image sorting unit, the corresponding mesh of each area monitoring image for identifying front-end image analytic unit
Logo image classification is sent to the back-end image analytic unit;
Back-end image analytic unit, for passing through the corresponding target identification parser of different target image category,
Analysis is identified to each area monitoring image, obtains the corresponding goal behavior information of each area monitoring image.
Further, the target image classification includes vehicle image, and/or pedestrian image, and/or parking lot environment map
Picture;The target identification parser includes algorithm of target detection, target tracking algorism, goal behavior parser;The mesh
It includes vehicle detecting algorithm, and/or pedestrian detection algorithm, and/or environment measuring algorithm to mark detection algorithm;The goal behavior point
It includes vehicle behavior parser, and/or pedestrian behavior parser, and/or ambient condition parser to analyse algorithm.
Further, the back-end image analytic unit, if being vehicle image specifically for the target image classification,
By vehicle detecting algorithm, target tracking algorism, vehicle behavior parser, analysis is identified to the vehicle image.
Further, the back-end image analytic unit, if it is pedestrian image to be specifically additionally operable to the target image classification,
Then by pedestrian detection algorithm, target tracking algorism, pedestrian behavior parser, analysis is identified to the pedestrian image.
Further, the back-end image analytic unit, if it is parking lot environment to be specifically additionally operable to the target image classification
Image carries out the parking lot ambient image then by environment measuring algorithm, target tracking algorism, ambient condition parser
Discriminance analysis.
Further, the event analysis module, it is right respectively according to synchronization associated region monitoring image to be specifically used for
The goal behavior information answered judges to whether there is abnormal object image in each area monitoring image.
Further, the event analysis module is specifically additionally operable to according to the same area monitoring image in preset time period
Corresponding goal behavior information judges to whether there is abnormal object image in each area monitoring image.
Further, the event analysis module is additionally operable to true according to the corresponding goal behavior information of abnormal object image
Determine the corresponding event type of abnormal object image.
Further, the decision configuration module is specifically used for the corresponding abnormality processing of configuration different event type
Strategy;
The abnormality processing executes equipment, is specifically used for according to the corresponding event type of the abnormal object image and decision
The corresponding abnormality processing strategy of different event type of configuration module configuration carries out corresponding abnormality processing.
Further, the abnormality processing executes equipment, is specifically used for region position corresponding in the abnormal object image
Set outputting alarm information;And/or solve prompt message in the corresponding regional location output abnormality of the abnormal object image;With/
Or send control instruction to the equipment of the corresponding regional location of the abnormal object image.
Further, the event type include vehicle traveling anomalous event, and/or pedestrian behavior anomalous event, and/or
Vehicle flowrate anomalous event, and/or parking lot environmental abnormality event.
Further, the abnormality processing executes equipment, if being specifically used for the corresponding event class of the abnormal object image
Type is that vehicle travels anomalous event, and/or pedestrian behavior anomalous event, and/or parking lot environmental abnormality event, then in the exception
The corresponding regional location outputting alarm information of target image.
Further, the abnormality processing executes equipment, if being specifically used for the corresponding event class of the abnormal object image
Type be vehicle flowrate anomalous event and/or parking lot environmental abnormality event, then it is defeated in the corresponding regional location of the abnormal object image
Go out abnormal solution prompt message.
Further, the abnormality processing executes equipment, if being specifically used for the corresponding event class of the abnormal object image
Type is parking lot environmental abnormality event and/or vehicle flowrate anomalous event, executes anomalous event processing control corresponding with event type
Instruct the operation of instruction.
Further, if the event type, which is vehicle, travels anomalous event,
Described image analysis module is additionally operable to identify the corresponding vehicle identification information of the abnormal object image;And by institute
Vehicle identification information is stated to be stored in image storage module.
Further, if the event type is pedestrian behavior anomalous event,
Described image analysis module is additionally operable to identify car owner's characteristic information in the abnormal object image;And it will be described
Car owner's characteristic information is stored in image storage module.
A kind of parking lot order management method and system based on image provided by the invention, by active to collected
Monitoring image carries out analysis and identification, and automatic identification abnormal object image may be implemented and monitor the abnormal feelings in each region in parking lot
Condition can support to show abnormal object image automatically, and automatic orientation abnormal monitoring is dredged, the work(such as automatic early-warning automatically
Can, to improve parking lot management efficiency and abnormal judging nicety rate, reduce cost of labor.
Description of the drawings
Fig. 1 is a kind of flow chart of parking lot order management method based on image provided by the invention;
Fig. 2 is a kind of structural schematic diagram of parking lot order management system based on image provided by the invention;
Fig. 3 is the structure of image analysis module in a kind of parking lot order management system based on image provided by the invention
Schematic diagram.
Specific implementation mode
Below by drawings and examples, technical scheme of the present invention will be described in further detail.
As shown in Figure 1, the embodiment of the present invention provides a kind of parking lot order management method based on image, including walk as follows
Suddenly:
101, each area monitoring image in parking lot is acquired.
Wherein, each area monitoring image in parking lot includes the vehicles while passing image information of gate area, pedestrian image letter
Breath, entrance equipment image information, foreign matter image information;Vehicle image information, pedestrian image information in the channel of parking lot, foreign matter
Image information;Vehicle image information, pedestrian image information, foreign matter image information of parking area etc., specifically can be by each
The video camera of a region arrangement carries out the acquisition of respective image information, and foreign matter image information can be rubbish image information, animal
Image information etc., the embodiment of the present invention does not limit.
102, analysis is identified to each area monitoring image, obtains the corresponding target of each area monitoring image
Behavioural information.
For the embodiment of the present invention, step 102 can specifically include:Identify the corresponding target of each area monitoring image
Image category specifically can carry out target image classification identification according to the type and attribute of video camera, then pass through different target
The corresponding target identification parser of image category is identified analysis to each area monitoring image, obtains each area
The corresponding goal behavior information of domain monitoring image.For the embodiment of the present invention, by identifying that each area monitoring image is distinguished
Corresponding target image classification and by the corresponding target identification parser of different target image category, to each area
Domain monitoring image carries out initiative recognition analysis, realizes and carries out High Precision Automatic identification function to the target in monitoring image.
Wherein, target image classification includes vehicle image, and/or pedestrian image, and/or parking lot ambient image, parking lot ring
Border image may include parking lot lighting condition image, parking lot equipment working state image etc..Target identification parser includes mesh
Mark detection algorithm, target tracking algorism, goal behavior parser, the algorithm of target detection include vehicle detecting algorithm and/
Or pedestrian detection algorithm, and/or environment measuring algorithm;Target tracking algorism can be the duscriminant image based on correlation computations with
Track algorithm (KCF, Kernelized Correlation Filters), the image tracking algorithm based on deep learning, Kalman
The image tracking algorithm etc. of filtering;The goal behavior parser includes vehicle behavior parser, and/or pedestrian behavior point
Analyse algorithm, and/or ambient condition parser.
Classification based on above-mentioned target image classification and target identification parser, to each area monitoring image into
Row discriminance analysis can specifically include following three kinds of realization methods but not limited to this:If mode one, the target image classification are
Vehicle image carries out the vehicle image then by vehicle detecting algorithm, target tracking algorism, vehicle behavior parser
Discriminance analysis.If mode two, the target image classification be pedestrian image, by pedestrian detection algorithm, target tracking algorism,
Analysis is identified to the pedestrian image in pedestrian behavior parser.If mode three, the target image classification are parking lot ring
Border image, then by environment measuring algorithm, target tracking algorism, ambient condition parser, to the parking lot ambient image into
Row discriminance analysis.For the embodiment of the present invention, first by algorithm of target detection lock onto target, then pass through target tracking algorism
To the dynamic of target into line trace, goal behavior information is obtained finally by goal behavior parser, to realize to prison
The initiative recognition analysis for controlling image, improves parking lot management efficiency.
Wherein, environment measuring algorithm is specifically as follows fire defector algorithm, Smoke Detection algorithm etc..Vehicle behavior analysis is calculated
Method is specifically as follows parking offense behavioural analysis algorithm, vehicle congestion behavioural analysis algorithm, car accident behavioural analysis algorithm etc.;
Pedestrian behavior parser is specifically as follows pedestrian and is detained behavioural analysis algorithm;Ambient condition parser is specifically as follows equipment
Abnormal behavior parser, fire alarm parser etc., the embodiment of the present invention does not limit.
103, judged in each area monitoring image according to the corresponding goal behavior information of each area monitoring image
With the presence or absence of abnormal object image.
Wherein, abnormal object image be image in include abnormal object behavioural information image.Abnormal object behavior is believed
Breath can be to occupy driving channel, vehicle in entrance delay, crowd massing, object blocking etc..
Specifically, judge that the mode of abnormal object image can specifically include following two modes but unlimited in step 103
In this:Mode one judges each region according to the corresponding goal behavior information of synchronization associated region monitoring image
It whether there is abnormal object image in monitoring image.Wherein, associated region can be track homonymy adjacent area, track to lateral areas
There are the multiple regions of interrelated relationship, the embodiment of the present invention does not limit for domain, parking lot exit and entry region etc..
Mode two judges institute according to the corresponding goal behavior information of the same area monitoring image in preset time period
It states and whether there is abnormal object image in each area monitoring image.Wherein, preset time period can be matched according to actual demand
It sets, such as 30 seconds, 1 minute, 30 minutes etc., the embodiment of the present invention does not limit.It specifically can be according to the same area preset time
The corresponding goal behavior information of different moments monitoring image in section judges in each area monitoring image with the presence or absence of different
Normal target image, above-mentioned different moments are each moment there are event correlation relationship, pedestrian's feature in the image such as taken
Information is identical or vehicle identification information identical each moment.For the embodiment of the present invention, pass through the above-mentioned association of comprehensive analysis
Region or the corresponding goal behavior information of correlation time monitoring image, to carry out abnormal object image and event type really
Recognize, can further promote the accuracy rate of abnormal object picture control and event type recognition.
For the concrete application scene of mode one, for example, in some area monitoring image taking to the vehicle stagnated, then
The image for transferring multiple continuous video cameras of the synchronization regional deployment adjacent with the region at this time, if having taken stagnation
Vehicle image, then may determine that in the area monitoring image and there are abnormal object figures in adjacent area monitoring image
Picture, and can judge that the corresponding event type of abnormal object image is vehicle congestion event according to this.
For the concrete application scene of mode two, for another example a certain within some preset time period, such as in 1 minute
A region each second takes same vehicle and is in outside berth always, then can determine that the monitoring image is abnormal object figure at this time
Picture, and judge that the corresponding event type of abnormal object image is vehicle peccancy Parking according to this.
104, when judgement is there are when abnormal object image, the abnormal object image is highlighted and carries out exception
Processing.
Wherein, it can be to be highlighted, amplify display, full frame to carry out highlighted mode to the abnormal object image
Modes, the embodiment of the present invention such as display abnormal object image do not limit.The embodiment of the present invention exists abnormal by initiative recognition
The monitoring image of target, and the monitoring image is highlighted, realize the automation supervision of monitoring image and abnormal mesh
Logo image is shown, so as to quickly find abnormal conditions existing for parking lot, avoids staff because needing while checking more
A monitoring area picture and the multiple monitoring area pictures of continuous poll is needed to find anomalous events and cause inefficient and eye
Fatigue causes the problem of anomalous event missing inspection, and then can improve the efficiency of management and management precision in parking lot.
For the embodiment of the present invention, before abnormality processing can also include the following steps:According to abnormal object image
Corresponding goal behavior information determines the corresponding event type of abnormal object image.
For example, the corresponding target image classification of monitoring image is parking lot ambient image, goal behavior information is the region vehicle
Field environment is dark, then can determine that the area monitoring image is abnormal object image, and determine the corresponding thing of abnormal object image
Part type is lighting failure.For another example the corresponding target image classification of monitoring image is parking lot ambient image, goal behavior
Information is that parking lot entrance banister is closed always, then can determine that the area monitoring image is abnormal object figure
Picture, and determine that the corresponding event type of abnormal object image is entrance banister failure.For another example the corresponding target of monitoring image
Image category is vehicle image, the goal behavior information identified be stoppage of vehicle among two berths, then can determine
The area monitoring image is abnormal object image, and determines that the corresponding event type of abnormal object image stops for vehicle peccancy.
For another example the corresponding target image classification of monitoring image is pedestrian image, the goal behavior information identified goes out for parking lot
The a large amount of gathering of people of mouth can then determine that the area monitoring image is abnormal object image, and determine that abnormal object image corresponds to
Event type be parking lot entrance crowd massing.Based on the corresponding event class of abnormal object image determined in above-mentioned steps
Type, abnormality processing is specifically as follows carries out corresponding abnormality processing according to the corresponding event type of abnormal object image.
Wherein, abnormality processing strategy can be refined as following three kinds of modes but not limited to this:In the abnormal object image
Corresponding regional location outputting alarm information, for playing the role of alarm to abnormal behaviour vehicle and abnormal behaviour pedestrian;Institute
State the corresponding regional location output abnormality of abnormal object image and solve prompt message, for congestion event, car accident event,
Indicative function is dredged in the offer of the events such as parking lot natural accident;It is sent out to the equipment of the corresponding regional location of the abnormal object image
Control instruction is sent, for carrying out automation control to equipment in parking lot, to realize at the correlation to parking lot anomalous event
Understanding is determined.The embodiment of the present invention by above-mentioned anomalous event processing mode, may be implemented parking lot order manage automatically, parking lot
The function that accident actively solves.The event type include vehicle traveling anomalous event, and/or pedestrian behavior anomalous event and/
Or vehicle flowrate anomalous event, and/or parking lot environmental abnormality event.
Based on the mode classification of above-mentioned event, if the corresponding event type of the abnormal object image is vehicle, traveling is abnormal
Event, and/or pedestrian behavior anomalous event, and/or parking lot environmental abnormality event, then in the corresponding area of the abnormal object image
Domain position outputting alarm information.For example, when there are parking offense events in berth region, or there are the exceptions such as vehicle theft event
Event or there are when the contingencies such as fire can pass through the output sound polices such as the loud speaker, the buzzer that are configured in the region
It accuses, or the light alarm by being configured in the region carries out light warning, the embodiment of the present invention does not limit.
Further, if the corresponding event type of the abnormal object image is vehicle flowrate anomalous event and/or parking lot ring
Border anomalous event then solves prompt message in the corresponding regional location output abnormality of the abnormal object image.For example, working as parking lot
When fire or vehicle congestion event occur for a certain region, it can be prompted by light guides or the modes such as display screen is shown are evacuated
Crowd or vehicle.
Further, if the corresponding event type of the abnormal object image is parking lot environmental abnormality event and/or wagon flow
Anomalous event is measured, control instruction is sent to the equipment of the corresponding regional location of the abnormal object image.For example, when parking lot is a certain
When the parking lots environmental abnormality event such as fire or vehicle congestion event occurs for region, can by gate to the region, channel into
Row closing control reaches conflagration area isolation, water logging zone isolation, congestion regions forbid new vehicle to drive into, congestion regions vehicle
The purpose of evacuation, realizes vehicle and evacuating personnel so as to occur phenomena such as preventing the condition of a disaster from spreading or congestion level aggravation
Purpose.
For the embodiment of the present invention, when there are abnormal object images and abnormality processing execution equipment failure for judgement
When, related personnel can also be notified to carry out artificial treatment to scene by way of sending out notice message, i.e., by event information and
The corresponding location information of event is pushed to related personnel, ensures to remain able to realize when abnormality processing executes equipment failure different
Chang Hangwei processing further improves the reliability and compatibility of the parking lot order management method provided by the invention based on image
Property.
For the embodiment of the present invention, can also in abnormal object image vehicle identification information and car owner's characteristic information into
Row storage, formed parking lot abnormal object database, when have the abnormal vehicle identification information preserved with abnormal object database or
It is pre- to provide safety when either pedestrian is re-introduced into parking lot or needs to enter parking lot for the abnormal matched vehicle of car owner's characteristic information
Alert function, further promotes the safety of parking lot management.Wherein, vehicle identification information can be vehicle license plate information, vehicle
Factory serial number information etc., car owner's characteristic information can be human face image information, iris image information etc., and the embodiment of the present invention does not limit
It is fixed.
Parking lot order management method provided in an embodiment of the present invention based on image can be applied to parking lot entrance
Congestion management, parking lot internal channel management, parking lot pedestrian's abnormal behaviour management, stops at the management of parking lot entrance flow of the people
In the various parking lot order management scenes such as parking lot vehicle abnormality behavior management.Collected monitoring image is carried out by active
Analysis and identification may be implemented automatic identification abnormal object image and monitor the abnormal conditions in each region in parking lot, can support pair
Abnormal object image shown automatically, and automatic orientation abnormal monitoring is dredged, the functions such as automatic early-warning automatically, to improve
Parking lot management efficiency and abnormal judging nicety rate, reduce cost of labor.
The method that embodiment provides to realize the present invention, the embodiment of the present invention provide a kind of parking lot order based on image
Management system, as shown in Fig. 2, the system includes:Image collecting device 21, image storage module 22, image analysis module 23, thing
Part analysis module 24, decision configuration module 25, abnormality processing execute equipment 26, network communication module 27;
Image collecting device 21, for acquiring each area monitoring image in parking lot.
Wherein, image collecting device 21 includes entrance monitor camera, parking lot internal channel monitor camera, parking stall prison
Control video camera.All kinds of monitor cameras can be visible light camera, thermal camera etc..Entrance monitor camera packet
Containing an at least video camera, it is deployed in the suitable view position of near exit, gate area is monitored, acquires gate area
Vehicles while passing image information, pedestrian image information, entrance equipment image information, foreign matter image information etc..Parking lot internal channel
Monitor camera includes an at least video camera, is deployed in the suitable view position of driving passage proximate of inner part of parking lot, is monitored
Corresponding passage area, vehicle image information, pedestrian image information, foreign matter image information etc. in acquisition channel.Parking position monitoring
Video camera includes an at least video camera, is deployed in suitable view position near parking stall, monitors corresponding several parking areas, adopts
Collect vehicle image information, pedestrian image information, the foreign matter image information etc. of parking area.It is real by the deployment of multiclass video camera
Now to the conduct monitoring at all levels in parking lot.
Image storage module 22, each area monitoring image in parking lot for storing the acquisition of image collecting device 21.
Wherein, described image memory module 22 can be hard disk video recorder.By data line or network, Image Acquisition is received
The image information of all kinds of video cameras of module 21, and store onto storage medium, image storage module 22 passes through network communication mould
Block 27 provides the calling of image information.
Image analysis module 23, described in the storage of acquired to image collecting device 21 or image storage module 22
Analysis is identified in each area monitoring image, obtains the corresponding goal behavior information of each area monitoring image.
Event analysis module 24, it is described each for being judged according to the corresponding goal behavior information of each area monitoring image
It whether there is abnormal object image in area monitoring image.
Wherein, abnormal object image is that there are the images of abnormal object behavioural information.For example, the corresponding target of monitoring image
Image category is parking lot ambient image, and goal behavior information is the region parking lot dumping, then can determine the area monitoring
Image is abnormal object image, and determines that the corresponding event type of abnormal object image is that rubbish leaves about event.For another example monitoring
The corresponding target image classification of image is parking lot ambient image, and goal behavior information is in for parking lot entrance banister and opens always
State is opened, then can determine that the area monitoring image is abnormal object image, and determine the corresponding event class of abnormal object image
Type is entrance banister failure.For another example the corresponding target image classification of monitoring image is pedestrian image, the target identified
Behavioural information is to draw vehicle, pound vehicle etc. to destroy vehicle behavior, then can determine that the area monitoring image is abnormal object image, and really
It is artificial vehicle destructive insident to determine the corresponding event type of abnormal object image.
Decision configuration module 25, for arrangement abnormalities processing strategy.
Abnormality processing executes equipment 26, for when judgement is there are when abnormal object image, to the abnormal object image into
Row highlights and carries out abnormality processing according to the abnormality processing strategy that decision configuration module 25 configures.
Wherein, it can be the side such as to be highlighted, amplify display to carry out highlighted mode to the abnormal object image
Formula, the embodiment of the present invention do not limit.It can be indicator light, gate equipment, loud speaker, buzzing that abnormality processing, which executes equipment 26,
Device, headlamp, fire-alarm, guiding equipment etc., the embodiment of the present invention does not limit.The embodiment of the present invention passes through initiative recognition
There are the monitoring images of abnormal object, and are highlighted to the monitoring image, realize the automation supervision of monitoring image
It is shown with abnormal object image, so as to quickly find abnormal conditions existing for parking lot, the management in parking lot can be improved
Efficiency.
Network communication module 27, for supporting image collecting device, image storage module, image analysis module, event point
Analysis module, decision configuration module, abnormality processing carry out information exchange between executing equipment.
Wherein, network communication module 27 can be the network switch, router, optical network unit, Wireless Communication Equipment
Deng.
Further, the structural schematic diagram of image analysis module 23 can be with as shown in figure 3, include:Front-end image analysis is single
Member 31, image sorting unit 32, back-end image analytic unit 33.
Front-end image analytic unit 31, for identification corresponding target image classification of each area monitoring image.
Wherein, front-end image analytic unit 31 is deployed in parking lot, passes through data line or network and image capture module
21 corresponding video camera connections, receive the image information of several video cameras, and according to the type and attribute of specific video camera, analysis
Image information obtains target image classification, such as vehicle, pedestrian, abnormal object.It is also possible to be front-end image analytic unit
31 configure some simple target identification parsers, are tentatively identified to goal behavior information.Front-end image analytic unit
31 provide preliminary image recognition result, submit to image sorting unit 32, and then trigger back-end image analytic unit 33.Front end
Image analyzing unit 31 can be deployed on the control host in parking lot.The control host can be embedded host.This is embedding
Entering formula host can support the image procossing based on graphics processor (GPU, Graphics Processing Unit) to accelerate.It should
Embedded host can also be supported to be based on field programmable gate array (FPGA, Field-Programmable Gate Array)
Image procossing accelerate.
Image sorting unit 32, each area monitoring image for identifying front-end image analytic unit 31 correspond to respectively
Target image classification be sent to the back-end image analytic unit 33.
Wherein, image sorting unit 32 can be according to front-end image analytic unit 31 as a result, by corresponding image information
It is distributed to the different disposal property server of back-end image analytic unit 33, carries out targetedly intensive treatment.
Back-end image analytic unit 33, for being calculated by the corresponding target identification analysis of different target image category
Method is identified analysis to each area monitoring image, obtains the corresponding goal behavior information of each area monitoring image.
Wherein, target image classification includes vehicle image, and/or pedestrian image, and/or parking lot ambient image, parking lot ring
Border image may include parking lot lighting condition image, parking lot equipment working state image etc..Target identification parser includes mesh
Mark detection algorithm, target tracking algorism, goal behavior parser, the algorithm of target detection include vehicle detecting algorithm and/
Or pedestrian detection algorithm, and/or environment measuring algorithm, the goal behavior parser include vehicle behavior parser and/
Or pedestrian behavior parser, and/or ambient condition parser.Back-end image analytic unit 33 can be executed based on depth
The algorithm of target detection and respective software module of habit, target tracking algorism and respective software module, behavioural analysis algorithm and correspondence
Software module.Back-end image analytic unit 33 can be processor cluster, can include that GPU accelerates cluster, can also include
FPGA clusters.For the embodiment of the present invention, back-end image analytic unit 33 can also be analyzed single in combination with multiple front-end images
The result and image information of member 31 carry out Conjoint Analysis, to further promote image recognition precision.
In embodiments of the present invention, environment measuring algorithm can be fire defector algorithm, Smoke Detection algorithm etc..Target with
Track algorithm can be the duscriminant image tracking algorithm based on correlation computations, the image tracking algorithm based on deep learning, karr
The image tracking algorithm etc. of graceful filtering.Vehicle behavior parser can be parking offense behavioural analysis algorithm, vehicle congestion row
For parser, car accident behavioural analysis algorithm etc.;Pedestrian behavior parser can be that pedestrian is detained behavioural analysis algorithm;
Ambient condition parser can be equipment behavior anomaly analysis algorithm, fire alarm parser etc., and the embodiment of the present invention is not
It limits.
Mode classification based on above-mentioned target image and above-mentioned each algorithm, the back-end image analytic unit 33, tool
If body is vehicle image for the target image classification, pass through vehicle detecting algorithm, target tracking algorism, vehicle behavior point
Algorithm is analysed, analysis is identified to the vehicle image;If specific be additionally operable to the target image classification as pedestrian image, lead to
Pedestrian's detection algorithm, target tracking algorism, pedestrian behavior parser are crossed, analysis is identified to the pedestrian image;Specifically
If it is parking lot ambient image to be additionally operable to the target image classification, pass through environment measuring algorithm, target tracking algorism, environment shape
Analysis is identified to the parking lot ambient image in state parser.
Further, event analysis module 24, specifically for being corresponded to respectively according to synchronization associated region monitoring image
Goal behavior information, judge in each area monitoring image whether there is abnormal object image.Or according to preset time
The corresponding goal behavior information of the same area monitoring image in section judges in each area monitoring image with the presence or absence of different
Normal target image.
Wherein, preset time period can be configured according to actual demand, such as 30 seconds, 1 minute, 30 minutes etc., this hair
Bright embodiment does not limit.For the embodiment of the present invention, pass through comprehensive analysis associated region or correlation time monitoring image pair
The goal behavior information answered can further promote abnormal object figure to carry out the confirmation of abnormal object image and event type
As the accuracy rate of monitoring and event type recognition.
Further, the event analysis module 24 is additionally operable to according to the corresponding goal behavior information of abnormal object image
Determine the corresponding event type of abnormal object image;At this point, the decision configuration module 25, is specifically used for configuration different event class
The corresponding abnormality processing strategy of type;Abnormality processing executes equipment 26, is specifically used for being corresponded to according to the abnormal object image
Event type and the corresponding abnormality processing strategy of different event type that configures of decision configuration module 25 carry out it is corresponding different
Often processing.
Wherein, the abnormality processing strategy can include but is not limited to such as under type:It is corresponded in the abnormal object image
Regional location outputting alarm information, for playing the role of alarm to abnormal behaviour vehicle and abnormal behaviour pedestrian;And/or institute
State the corresponding regional location output abnormality of abnormal object image and solve prompt message, for congestion event, car accident event,
Indicative function is dredged in the offer of the events such as parking lot natural accident;And/or to the corresponding regional location of the abnormal object image
Equipment sends control instruction to realize that the relevant treatment to parking lot anomalous event solves.The embodiment of the present invention passes through above-mentioned different
The function that parking lot order manages automatically, parking lot accident actively solves may be implemented in ordinary affair part processing mode.
For the embodiment of the present invention, the event type includes vehicle traveling anomalous event, and/or pedestrian behavior exception thing
Part, and/or vehicle flowrate anomalous event, and/or parking lot environmental abnormality event.
The mode classification of configuration mode and event based on above-mentioned abnormality processing strategy, the abnormality processing execute equipment
26, if being that vehicle travels anomalous event, and/or pedestrian behavior is different specifically for the corresponding event type of the abnormal object image
Ordinary affair part, and/or parking lot environmental abnormality event, then in the corresponding regional location outputting alarm information of the abnormal object image;
If the corresponding event type of the abnormal object image is vehicle flowrate anomalous event and/or parking lot environmental abnormality event, in institute
It states the corresponding regional location output abnormality of abnormal object image and solves prompt message;If the corresponding event of the abnormal object image
Type is parking lot environmental abnormality event and/or vehicle flowrate anomalous event, executes decision configuration module 25 configures and event type
The operation of corresponding anomalous event processing control instruction instruction.
Further, if the corresponding event type of the abnormal object image is parking lot environmental abnormality event, image analysis
Module 23 is additionally operable to identify the corresponding vehicle identification information of the abnormal object image;And the vehicle identification information is stored in
In image storage module 22.
Further, if the event type is pedestrian behavior anomalous event, image analysis module 23 is additionally operable to identification institute
State car owner's characteristic information in abnormal object image;And car owner's characteristic information is stored in image storage module 22.
Wherein, vehicle identification information can be vehicle license plate information, vehicle release information etc., and car owner's characteristic information can be with
For human face image information, iris image information etc., the embodiment of the present invention does not limit.For the embodiment of the present invention, by different
Vehicle identification information and car owner's characteristic information in normal target image store, and form parking lot abnormal object database, when having
The abnormal vehicle identification information preserved with abnormal object database the either matched vehicle of abnormal car owner's characteristic information or pedestrian
When being re-introduced into parking lot or needing to enter parking lot, safety precaution function is provided, further promotes the safety of parking lot management
Property.
Other explanations for each device in system provided in an embodiment of the present invention, can refer to the embodiment of the present invention
The corresponding of each step illustrates in the method for offer, and details are not described herein.
Parking lot order management system provided by the invention based on image can be applied to parking lot entrance congestion pipe
The various parkings such as reason, parking lot internal channel management, parking lot pedestrian's abnormal behaviour management, parking lot vehicle abnormality behavior management
In the order management scene of field.Analysis and identification is carried out to collected monitoring image by active, automatic identification exception may be implemented
The abnormal conditions of target image and the monitoring each region in parking lot, can support to show abnormal object image automatically, automatically
Orientation abnormal monitoring is dredged, the functions such as automatic early-warning automatically, to improve parking lot management efficiency and exception accuracy of judgement
Rate reduces cost of labor.
Above-described specific implementation mode has carried out further the purpose of the present invention, technical solution and advantageous effect
It is described in detail, it should be understood that the foregoing is merely the specific implementation mode of the present invention, is not intended to limit the present invention
Protection domain, all within the spirits and principles of the present invention, any modification, equivalent substitution, improvement and etc. done should all include
Within protection scope of the present invention.
Claims (34)
1. a kind of parking lot order management method based on image, which is characterized in that the method includes:
Acquire each area monitoring image in parking lot;
Analysis is identified to each area monitoring image, obtains the corresponding goal behavior letter of each area monitoring image
Breath;
Judge to whether there is in each area monitoring image according to the corresponding goal behavior information of each area monitoring image
Abnormal object image;
When judgement is there are when abnormal object image, the abnormal object image is highlighted and carries out abnormality processing.
2. the parking lot order management method according to claim 1 based on image, which is characterized in that described to described each
Analysis is identified in area monitoring image, and the step of obtaining each area monitoring image corresponding goal behavior information includes:
Identify the corresponding target image classification of each area monitoring image;
By the corresponding target identification parser of different target image category, each area monitoring image is known
It does not analyze, obtains the corresponding goal behavior information of each area monitoring image.
3. the parking lot order management method according to claim 2 based on image, which is characterized in that the target image
Classification includes vehicle image, and/or pedestrian image, and/or parking lot ambient image;
The target identification parser includes algorithm of target detection, target tracking algorism, goal behavior parser;The mesh
It includes vehicle detecting algorithm, and/or pedestrian detection algorithm, and/or environment measuring algorithm to mark detection algorithm;The goal behavior point
It includes vehicle behavior parser, and/or pedestrian behavior parser, and/or ambient condition parser to analyse algorithm.
4. the parking lot order management method based on image according to right 3, which is characterized in that pass through different target image
The corresponding target identification parser of classification, the step of analysis is identified to each area monitoring image include:
If the target image classification is vehicle image, analyzed by vehicle detecting algorithm, target tracking algorism, vehicle behavior
Analysis is identified to the vehicle image in algorithm.
5. the parking lot order management method based on image according to right 3, which is characterized in that pass through different target image
The corresponding target identification parser of classification, the step of analysis is identified to each area monitoring image include:
If the target image classification is pedestrian image, analyzed by pedestrian detection algorithm, target tracking algorism, pedestrian behavior
Analysis is identified to the pedestrian image in algorithm.
6. the parking lot order management method based on image according to right 3, which is characterized in that pass through different target image
The corresponding target identification parser of classification, the step of analysis is identified to each area monitoring image include:
If the target image classification is parking lot ambient image, pass through environment measuring algorithm, target tracking algorism, ambient condition
Analysis is identified to the parking lot ambient image in parser.
7. the parking lot order management method according to claim 1 based on image, which is characterized in that supervised according to each region
The corresponding goal behavior information of control image judges to whether there is the step of abnormal object image in each area monitoring image
Suddenly include:
According to the corresponding goal behavior information of synchronization associated region monitoring image, each area monitoring image is judged
In whether there is abnormal object image.
8. the parking lot order management method according to claim 1 based on image, which is characterized in that supervised according to each region
The corresponding goal behavior information of control image judges to whether there is the step of abnormal object image in each area monitoring image
Suddenly include:
According to the corresponding goal behavior information of the same area monitoring image in preset time period, each area monitoring is judged
It whether there is abnormal object image in image.
9. the parking lot order management method according to claim 1 based on image, which is characterized in that the progress is abnormal
Before the step of processing, the method further includes:
The corresponding event type of abnormal object image is determined according to the corresponding goal behavior information of abnormal object image.
10. the parking lot order management method according to claim 9 based on image, which is characterized in that the progress is different
Often processing the step of include:
Corresponding abnormality processing is carried out according to the corresponding event type of the abnormal object image.
11. the parking lot order management method according to claim 10 based on image, which is characterized in that the exception
Managing strategy includes:
In the corresponding regional location outputting alarm information of the abnormal object image;And/or
Prompt message is solved in the corresponding regional location output abnormality of the abnormal object image;And/or
Control instruction is sent to the equipment of the corresponding regional location of the abnormal object image.
12. the parking lot order management method according to claim 11 based on image, which is characterized in that the event class
Type includes vehicle traveling anomalous event, and/or pedestrian behavior anomalous event, and/or vehicle flowrate anomalous event, and/or parking lot ring
Border anomalous event.
13. the parking lot order management method based on image according to right 12, which is characterized in that described according to described different
The step of corresponding event type of normal target image carries out corresponding abnormality processing include:
If the corresponding event type of the abnormal object image is vehicle travel anomalous event, and/or pedestrian behavior anomalous event,
And/or parking lot environmental abnormality event, then in the corresponding regional location outputting alarm information of the abnormal object image.
14. the parking lot order management method based on image according to right 12, which is characterized in that described according to described different
The step of corresponding event type of normal target image carries out corresponding abnormality processing include:
If the corresponding event type of the abnormal object image is vehicle flowrate anomalous event and/or parking lot environmental abnormality event,
Prompt message is solved in the corresponding regional location output abnormality of the abnormal object image.
15. the parking lot order management method based on image according to right 12, which is characterized in that described according to described different
The step of corresponding event type of normal target image carries out corresponding abnormality processing include:
If the corresponding event type of the abnormal object image is parking lot environmental abnormality event and/or vehicle flowrate anomalous event, to
The equipment of the corresponding regional location of the abnormal object image sends control instruction.
16. the parking lot order management method according to claim 12 based on image, which is characterized in that if the event
Type is that vehicle travels anomalous event, and the method further includes:
Identify the corresponding vehicle identification information of the abnormal object image;
The vehicle identification information is stored in abnormal vehicle information database.
17. the parking lot order management method according to claim 12 based on image, which is characterized in that if the event
Type is pedestrian behavior anomalous event, and the method further includes:
Identify car owner's characteristic information in the abnormal object image;
Car owner's characteristic information is stored in abnormal owner information database.
18. a kind of parking lot order based on image manages system, which is characterized in that the system comprises:Image collecting device,
Image storage module, image analysis module, event analysis module, decision configuration module, abnormality processing execute equipment, network communication
Module;
Image collecting device, for acquiring each area monitoring image in parking lot;
Image storage module, each area monitoring image in parking lot for storing image acquisition device;
Image analysis module, for image acquisition device or image storage module storage each area monitoring
Analysis is identified in image, obtains the corresponding goal behavior information of each area monitoring image;
Event analysis module, for judging that each region is supervised according to the corresponding goal behavior information of each area monitoring image
It controls and whether there is abnormal object image in image;
Decision configuration module, for arrangement abnormalities processing strategy;
Abnormality processing executes equipment, for when judgement is there are when abnormal object image, protruding the abnormal object image
It shows and abnormality processing is carried out according to the abnormality processing strategy of decision configuration module configuration;
Network communication module, for support image collecting device, image storage module, image analysis module, event analysis module,
Decision configuration module, abnormality processing carry out information exchange between executing equipment.
19. the parking lot order according to claim 18 based on image manages system, which is characterized in that described image point
Analysing module includes:Front-end image analytic unit, image sorting unit, back-end image analytic unit;
Front-end image analytic unit, for identification corresponding target image classification of each area monitoring image;
Image sorting unit, the corresponding target figure of each area monitoring image for identifying front-end image analytic unit
As classification is sent to the back-end image analytic unit;
Back-end image analytic unit, for passing through the corresponding target identification parser of different target image category, to institute
It states each area monitoring image and analysis is identified, obtain the corresponding goal behavior information of each area monitoring image.
20. the parking lot order according to claim 19 based on image manages system, which is characterized in that the target figure
As classification includes vehicle image, and/or pedestrian image, and/or parking lot ambient image;The target identification parser includes mesh
Mark detection algorithm, target tracking algorism, goal behavior parser;The algorithm of target detection include vehicle detecting algorithm and/
Or pedestrian detection algorithm, and/or environment measuring algorithm;The goal behavior parser include vehicle behavior parser and/
Or pedestrian behavior parser, and/or ambient condition parser.
21. the parking lot order according to claim 20 based on image manages system, which is characterized in that
The back-end image analytic unit passes through vehicle detection if being vehicle image specifically for the target image classification
Analysis is identified to the vehicle image in algorithm, target tracking algorism, vehicle behavior parser.
22. the parking lot order according to claim 20 based on image manages system, which is characterized in that
The back-end image analytic unit is examined if it is pedestrian image to be specifically additionally operable to the target image classification by pedestrian
Analysis is identified to the pedestrian image in method of determining and calculating, target tracking algorism, pedestrian behavior parser.
23. the parking lot order according to claim 20 based on image manages system, which is characterized in that
The back-end image analytic unit passes through ring if it is parking lot ambient image to be specifically additionally operable to the target image classification
Analysis is identified to the parking lot ambient image in border detection algorithm, target tracking algorism, ambient condition parser.
24. the parking lot order according to claim 18 based on image manages system, which is characterized in that
The event analysis module is specifically used for being believed according to the corresponding goal behavior of synchronization associated region monitoring image
Breath judges to whether there is abnormal object image in each area monitoring image.
25. the parking lot order according to claim 18 based on image manages system, which is characterized in that
The event analysis module is specifically additionally operable to according to the corresponding target of the same area monitoring image in preset time period
Behavioural information judges to whether there is abnormal object image in each area monitoring image.
26. the parking lot order according to claim 18 based on image manages system, which is characterized in that
The event analysis module is additionally operable to determine abnormal object image according to the corresponding goal behavior information of abnormal object image
Corresponding event type.
27. the parking lot order according to claim 26 based on image manages system, which is characterized in that
The decision configuration module is specifically used for the corresponding abnormality processing strategy of configuration different event type;
The abnormality processing executes equipment, is specifically used for being configured according to the corresponding event type of the abnormal object image and decision
The corresponding abnormality processing strategy of different event type of module configuration carries out corresponding abnormality processing.
28. the parking lot order according to claim 27 based on image manages system, which is characterized in that the exception
Managing strategy includes:
In the corresponding regional location outputting alarm information of the abnormal object image;And/or
Prompt message is solved in the corresponding regional location output abnormality of the abnormal object image;And/or
Control instruction is sent to the equipment of the corresponding regional location of the abnormal object image.
29. the parking lot order according to claim 28 based on image manages system, which is characterized in that the event class
Type includes vehicle traveling anomalous event, and/or pedestrian behavior anomalous event, and/or vehicle flowrate anomalous event, and/or parking lot ring
Border anomalous event.
30. the parking lot order according to claim 29 based on image manages system, which is characterized in that
The abnormality processing executes equipment, if being that vehicle traveling is different specifically for the corresponding event type of the abnormal object image
Ordinary affair part, and/or pedestrian behavior anomalous event, and/or parking lot environmental abnormality event, then it is corresponding in the abnormal object image
Regional location outputting alarm information.
31. the parking lot order according to claim 29 based on image manages system, which is characterized in that
The abnormality processing executes equipment, if being that vehicle flowrate is abnormal specifically for the corresponding event type of the abnormal object image
Event and/or parking lot environmental abnormality event then solve prompt in the corresponding regional location output abnormality of the abnormal object image
Information.
32. the parking lot order according to claim 29 based on image manages system, which is characterized in that
The abnormality processing executes equipment, if being that parking lot environment is different specifically for the corresponding event type of the abnormal object image
Ordinary affair part and/or vehicle flowrate anomalous event execute the operation of anomalous event processing control instruction instruction corresponding with event type.
33. the parking lot order according to claim 29 based on image manages system, which is characterized in that if the event
Type is that vehicle travels anomalous event,
Described image analysis module is additionally operable to identify the corresponding vehicle identification information of the abnormal object image;And by the vehicle
Identification information is stored in image storage module.
34. the parking lot order according to claim 29 based on image manages system, which is characterized in that if the event
Type is pedestrian behavior anomalous event,
Described image analysis module is additionally operable to identify car owner's characteristic information in the abnormal object image;And by the car owner
Characteristic information is stored in image storage module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810341890.0A CN108717521A (en) | 2018-04-17 | 2018-04-17 | A kind of parking lot order management method and system based on image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810341890.0A CN108717521A (en) | 2018-04-17 | 2018-04-17 | A kind of parking lot order management method and system based on image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108717521A true CN108717521A (en) | 2018-10-30 |
Family
ID=63899162
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810341890.0A Pending CN108717521A (en) | 2018-04-17 | 2018-04-17 | A kind of parking lot order management method and system based on image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108717521A (en) |
Cited By (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110097722A (en) * | 2019-03-20 | 2019-08-06 | 合肥名德光电科技股份有限公司 | Vehicle monitoring alarm method and device based on image procossing |
CN110136479A (en) * | 2019-06-13 | 2019-08-16 | 北京中镖国际保安服务有限公司 | Parking lot emergent treatment system |
CN110751839A (en) * | 2019-10-23 | 2020-02-04 | 深圳市盛世基业物联网有限公司 | Traffic management system in parking lot |
CN111277795A (en) * | 2020-01-15 | 2020-06-12 | 重庆特斯联智慧科技股份有限公司 | Parking lot monitoring method and system based on artificial intelligence |
CN111294554A (en) * | 2018-12-10 | 2020-06-16 | 丰田自动车株式会社 | Behavior monitoring device, behavior monitoring system, and behavior monitoring program |
CN112329499A (en) * | 2019-08-05 | 2021-02-05 | 上海高德威智能交通系统有限公司 | Image processing method, device and equipment |
CN112365740A (en) * | 2020-11-30 | 2021-02-12 | 北京停简单信息技术有限公司 | Alarm display method and device |
CN112967467A (en) * | 2021-02-24 | 2021-06-15 | 九江学院 | Cultural relic anti-theft method, system, mobile terminal and storage medium |
CN114882734A (en) * | 2022-07-12 | 2022-08-09 | 车位管家(深圳)科技有限公司 | Parking lot safety monitoring management system based on artificial intelligence |
TWI802168B (en) * | 2021-12-23 | 2023-05-11 | 兆竑智聯股份有限公司 | Abnormal detection method and system for security of parking lot |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101266710A (en) * | 2007-03-14 | 2008-09-17 | 中国科学院自动化研究所 | An all-weather intelligent video analysis monitoring method based on a rule |
CN101389004A (en) * | 2007-09-13 | 2009-03-18 | 中国科学院自动化研究所 | Moving target classification method based on on-line study |
CN101622652A (en) * | 2007-02-08 | 2010-01-06 | 行为识别系统公司 | The behavior recognition system |
US20110050899A1 (en) * | 2008-01-15 | 2011-03-03 | Marcel Merkel | Surveillance module for a video surveillance system, method for monitoring the state of a surveillance region, and computer program |
CN102164270A (en) * | 2011-01-24 | 2011-08-24 | 浙江工业大学 | Intelligent video monitoring method and system capable of exploring abnormal events |
KR101582254B1 (en) * | 2014-07-09 | 2016-01-04 | 목원대학교 산학협력단 | Underground parking securety system using cctv camera and mobil terminal, provide method thereof |
US20160283799A1 (en) * | 2013-04-19 | 2016-09-29 | James Carey | Video identification and analytical recognition system |
-
2018
- 2018-04-17 CN CN201810341890.0A patent/CN108717521A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101622652A (en) * | 2007-02-08 | 2010-01-06 | 行为识别系统公司 | The behavior recognition system |
CN101266710A (en) * | 2007-03-14 | 2008-09-17 | 中国科学院自动化研究所 | An all-weather intelligent video analysis monitoring method based on a rule |
CN101389004A (en) * | 2007-09-13 | 2009-03-18 | 中国科学院自动化研究所 | Moving target classification method based on on-line study |
US20110050899A1 (en) * | 2008-01-15 | 2011-03-03 | Marcel Merkel | Surveillance module for a video surveillance system, method for monitoring the state of a surveillance region, and computer program |
CN102164270A (en) * | 2011-01-24 | 2011-08-24 | 浙江工业大学 | Intelligent video monitoring method and system capable of exploring abnormal events |
US20160283799A1 (en) * | 2013-04-19 | 2016-09-29 | James Carey | Video identification and analytical recognition system |
KR101582254B1 (en) * | 2014-07-09 | 2016-01-04 | 목원대학교 산학협력단 | Underground parking securety system using cctv camera and mobil terminal, provide method thereof |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111294554A (en) * | 2018-12-10 | 2020-06-16 | 丰田自动车株式会社 | Behavior monitoring device, behavior monitoring system, and behavior monitoring program |
CN110097722A (en) * | 2019-03-20 | 2019-08-06 | 合肥名德光电科技股份有限公司 | Vehicle monitoring alarm method and device based on image procossing |
CN110136479A (en) * | 2019-06-13 | 2019-08-16 | 北京中镖国际保安服务有限公司 | Parking lot emergent treatment system |
CN112329499A (en) * | 2019-08-05 | 2021-02-05 | 上海高德威智能交通系统有限公司 | Image processing method, device and equipment |
CN110751839A (en) * | 2019-10-23 | 2020-02-04 | 深圳市盛世基业物联网有限公司 | Traffic management system in parking lot |
CN111277795A (en) * | 2020-01-15 | 2020-06-12 | 重庆特斯联智慧科技股份有限公司 | Parking lot monitoring method and system based on artificial intelligence |
CN112365740A (en) * | 2020-11-30 | 2021-02-12 | 北京停简单信息技术有限公司 | Alarm display method and device |
CN112967467A (en) * | 2021-02-24 | 2021-06-15 | 九江学院 | Cultural relic anti-theft method, system, mobile terminal and storage medium |
CN112967467B (en) * | 2021-02-24 | 2022-07-29 | 九江学院 | Cultural relic anti-theft method, system, mobile terminal and storage medium |
TWI802168B (en) * | 2021-12-23 | 2023-05-11 | 兆竑智聯股份有限公司 | Abnormal detection method and system for security of parking lot |
CN114882734A (en) * | 2022-07-12 | 2022-08-09 | 车位管家(深圳)科技有限公司 | Parking lot safety monitoring management system based on artificial intelligence |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108717521A (en) | A kind of parking lot order management method and system based on image | |
CN108537190A (en) | A kind of parking lot order managing device based on image | |
CN103108159B (en) | Electric power intelligent video analyzing and monitoring system and method | |
CN111325977A (en) | Tunnel intelligence edge calculation management and control system | |
CN106205143A (en) | Intelligent bayonet networked system | |
CN113256924A (en) | Monitoring system, monitoring method and monitoring device for rail train | |
CN103873825A (en) | ATM (automatic teller machine) intelligent monitoring system and method | |
CN105469035A (en) | Driver's bad driving behavior detection system based on binocular video analysis | |
CA2925843C (en) | System for traffic behaviour surveillance | |
US11386763B2 (en) | Luminaire network with sensors | |
CN105894700A (en) | Image-based in-vehicle moving object remote observing and warning device and method | |
CN105607617A (en) | Security fault diagnosis system and method based on Internet of Things | |
CN109389794A (en) | A kind of Intellectualized Video Monitoring method and system | |
JP2007195130A (en) | Video monitoring system | |
CN107464416B (en) | Semi-automatic driving method and system for bus | |
CN106454331A (en) | A video signal quality detection system and method | |
CN102493845A (en) | Monitoring and prewarning system for urban rail transit operating tunnel based on GIS (Geographic Information System) | |
CN112804489B (en) | Intelligent construction site management system and method based on Internet + | |
CN211630273U (en) | Intelligent image recognition device for railway environment | |
KR101729985B1 (en) | Tunnel integrated disaster prevention and broadcasting system for maintaining many tunnel integrated disaster prevention devices | |
CN209842394U (en) | Intelligent monitoring system for rail-side machine room of 5T equipment | |
CN110930707A (en) | Non-motor vehicle traffic violation supervision system and supervision method | |
CN113393625A (en) | Anti-theft alarm evidence obtaining method and system for intelligent lamp pole | |
CN113691778A (en) | Panoramic station patrol system for urban rail transit station | |
CN105872486A (en) | Video monitoring system with automatic image discriminating, analyzing and distinguishing device and application method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 075000 ten building, phase 1, airport economic and Technological Development Zone, Zhangjiakou, Hebei Applicant after: Smart intercommunication Technology Co.,Ltd. Address before: 075000 ten building, phase 1, airport economic and Technological Development Zone, Zhangjiakou, Hebei Applicant before: INTELLIGENT INTERCONNECTION TECHNOLOGIES Co.,Ltd. |