CN108305466A - Roadside Parking detection method and device based on vehicle characteristics analysis - Google Patents
Roadside Parking detection method and device based on vehicle characteristics analysis Download PDFInfo
- Publication number
- CN108305466A CN108305466A CN201810203056.5A CN201810203056A CN108305466A CN 108305466 A CN108305466 A CN 108305466A CN 201810203056 A CN201810203056 A CN 201810203056A CN 108305466 A CN108305466 A CN 108305466A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- recognition result
- video camera
- scene image
- parking
- 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.)
- Granted
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/292—Multi-camera tracking
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
- Devices For Checking Fares Or Tickets At Control Points (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides the Roadside Parking detection methods analyzed based on vehicle characteristics, including：Obtain the first continuous scene image；Doubtful parking area detection is carried out, parking stall coordinate information is obtained；Second video camera furthers parking stall coordinate information region automatically, carries out Car license recognition and vehicle inscriptions are other, obtains best Car license recognition and vehicle inscriptions not as a result, generating vehicle ID, line direction of going forward side by side detection；If direction is to drive into, the continuous scene image of third is obtained using the first video camera, vehicle ID track detections is carried out and stores vehicle ID if track does not update whithin a period of time, otherwise delete vehicle ID；If direction is to be driven out to, the 4th continuous scene image is obtained using the second video camera, carries out vehicle ID track detections, if the tracks vehicle ID are compared far from parking stall coordinate, by vehicle ID with the vehicle ID in database, exports result.Compared with prior art, the present invention can effectively solve the problems, such as license plate shading and unlicensed vehicle in Roadside Parking detection.
Description
Technical field
The present invention relates to image procossing, video monitoring and security protections, more particularly to Roadside Parking detection method and device.
Background technology
Increasing with car ownership, Roadside Parking also becomes one of urgent problem to be solved, has caused
The attention of vehicle supervision department.Traditional Roadside Parking detection is enclosure, ground-sensor and hand-held set, and efficiency is low, nothing
Method realizes real time monitoring, greatly wastes the manpower and financial resources of relevant departments.
In recent years, the method for carrying out Roadside Parking detection based on monitor video has been a great concern, and this method has
Accuracy rate is high, real-time is good, at low cost, the advantages that being easy to collect evidence.According to the position that monitoring camera is installed, there is the low bar of trackside
Scheme（0.8 ~ 1.8 meter of mounting height）With the high bar scheme of trackside（3.5 ~ 7 meters of mounting height）.Low bar scheme is pacified on a low bar
A video camera is filled, the state of a parking stall is monitored；High bar scheme is combined using multiple gun machine and more ball machines, and gunlock and ball are passed through
Machine linkage is realized multiple（2 ~ 12）The detection of parking stall.However, said program is all based on license plate recognition technology, therefore can not have
Effect solves license plate shading and unlicensed vehicle test problems.
In conclusion there is an urgent need to propose a kind of Roadside Parking detection that can effectively solve license plate shading and unlicensed vehicle at present
Method and device.
Invention content
In view of this, the main purpose of the present invention is to provide Roadside Parking detection scheme, car plate screening can be effectively solved
Gear and unlicensed vehicle problem.
In order to achieve the above objectives, the first aspect according to the invention is provided the trackside analyzed based on vehicle characteristics and stopped
Vehicle detection method, this method include：
First step obtains the first continuous scene image using the first video camera；
Second step carries out doubtful parking area detection to the first continuous scene image, and the parking stall for obtaining doubtful parking area is sat
Information is marked, and is sent to the second video camera；
Third step, the second video camera further parking stall coordinate information region automatically, obtain the second continuous scene image, carry out car plate
Identification and vehicle inscriptions are other, obtain best Car license recognition and vehicle inscriptions not as a result, generating vehicle ID, and carry out direction of motion detection；
Four steps sends out parking stall coordinate information, vehicle ID, direction of vehicle movement if the vehicle ID directions of motion are to drive into
The first video camera is given, the second video camera is returned to original state, and is transferred to the 5th step；If the vehicle ID directions of motion are
It is driven out to, is then transferred to the 6th step；
5th step obtains the continuous scene image of third using the first video camera, vehicle ID track detections is carried out, if vehicle ID
Track does not update within the second threshold time, then stores vehicle ID, parking stall coordinate information and down time and return, otherwise delete
Vehicle ID is simultaneously returned；
6th step obtains the 4th continuous scene image using the second video camera, carries out vehicle ID track detections, by the second camera shooting
Machine returns to original state, if the tracks vehicle ID are transferred to the 7th step, otherwise return far from parking stall coordinate；
Vehicle ID is compared 7th step with the vehicle ID in database, if compared successfully, records vehicle ID and stops
The relevant information of vehicle event；If comparing failure, records vehicle ID and compare failure information.
Wherein, returned described in the 5th step and the 6th step refers to returning to the first step to restart
Processing.
First video camera and the second video camera are mounted in trackside monitoring upright bar, and mounting height is located at ground 1.5 ~ 8
At rice, 2 ~ 12 parking stall ranges can be covered.
Further, first video camera is the video camera that focal length is not less than 3.6mm, and second video camera is zoom
Video camera.
Further, the second step includes：
Track of vehicle detecting step detects vehicle using the method for moving object detection and tracking according to the first continuous scene image
Movement locus；
Doubtful parking area detecting step, if the movement locus of vehicle is overlapping with a certain parking area, while vehicle enters friendship
The area of folded parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information exports step, and the corresponding parking stall coordinate information of doubtful parking area is sent to the second video camera.
Further, the third step includes：
Second continuous scene obtaining step, according to the parking stall coordinate information of the doubtful parking area of reception, the second video camera is automatic
It furthers the region of corresponding parking stall coordinate information, obtains the second continuous scene image；
Vehicle feature recognition step carries out Car license recognition to the second continuous scene image and vehicle inscriptions is other, obtains per frame image
Car license recognition and vehicle money recognition result；
The license plate recognition result of every frame image, vehicle money recognition result are merged, are obtained by multiframe recognition result fusion steps respectively
Take best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation steps, according to best license plate recognition result and optimized vehicle inscriptions not as a result, generating vehicle ID；
Direction of vehicle movement detecting step carries out track of vehicle detection, if vehicle to vehicle ID in the second continuous scene image
The track of vehicle of ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if the vehicle rail of vehicle ID
Mark is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
Further, the multiframe recognition result fusion steps include：It counts successively each in the second continuous scene image
The license plate recognition result of frame scene image calculates the quantity of identical license plate recognition result, chooses identical license plate recognition result quantity
License plate recognition result corresponding to maximum value, as best license plate recognition result；Count every in the second continuous scene image successively
The vehicle inscriptions of one frame scene image choose the identical other number of results of vehicle inscriptions not as a result, calculate the quantity of identical vehicle money recognition result
The vehicle inscriptions corresponding to maximum value are measured not as a result, as optimized vehicle money recognition result.
Further, the vehicle ID generation steps according to best license plate recognition result+optimized vehicle money recognition result shape
Formula generates vehicle ID.
Further, the 7th step includes：
Car plate compares step, calculates the matching degree of the license plate recognition result of the vehicle ID stored in vehicle ID and database, if
The matching degree of license plate recognition result is not less than third threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, otherwise turns
Enter vehicle money and compares step；
Vehicle money compares step, calculates the matching degree of the vehicle money recognition result of the vehicle ID stored in vehicle ID and database, if
The matching degree of vehicle money recognition result is not less than the 4th threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, otherwise turns
Enter car plate vehicle money Combined Ration to step；
Car plate vehicle money Combined Ration is to step, if the matching degree of license plate recognition result is not less than the 5th threshold value, and vehicle inscriptions are other
As a result matching degree is not less than the 6th threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, is lost otherwise it is assumed that comparing
It loses；
Comparison result exports step, if comparing the relevant information for successfully recording vehicle ID and Parking；It is lost if compared
It loses, then record vehicle ID and compares failure information.
Other side according to the invention provides the Roadside Parking detection device analyzed based on vehicle characteristics, the dress
Set including：
First continuous scene image acquisition module, for obtaining the first continuous scene image using the first video camera；
Doubtful parking area parking stall coordinate obtaining module, for carrying out doubtful parking area detection to the first continuous scene image,
The parking stall coordinate information of doubtful parking area is obtained, and is sent to the second video camera；
Vehicle ID and direction of motion acquisition module further parking stall coordinate information region automatically for the second video camera, obtain second
Continuous scene image, carries out Car license recognition and vehicle inscriptions are other, obtains best Car license recognition and vehicle inscriptions not as a result, generating vehicle
ID, and carry out direction of motion detection；
Selecting module based on the direction of motion, if being to drive into for the vehicle ID directions of motion, by parking stall coordinate information, vehicle
ID, direction of vehicle movement are sent to the first video camera, and the second video camera is returned to original state, and be transferred to vehicle ID storage and
Removing module；If the vehicle ID directions of motion are to be driven out to, it is transferred to vehicle and is driven out to processing module；
Vehicle ID storages and removing module carry out vehicle ID rails for obtaining the continuous scene image of third using the first video camera
Mark detects, if the tracks vehicle ID do not update within the second threshold time, when storing vehicle ID, parking stall coordinate information and parking
Between and return, otherwise delete and vehicle ID and return；
Vehicle is driven out to processing module, for obtaining the 4th continuous scene image using the second video camera, carries out the inspection of the tracks vehicle ID
It surveys, the second video camera is returned to original state, if the tracks vehicle ID are transferred to vehicle ID and compare mould far from parking stall coordinate
Otherwise block returns；
Vehicle ID comparing modules, if compared successfully, are recorded for vehicle ID to be compared with the vehicle ID in database
The relevant information of vehicle ID and Parking；If comparing failure, records vehicle ID and compare failure information.
Further, the doubtful parking area parking stall coordinate obtaining module includes：
Track of vehicle detection module, for the method using moving object detection and tracking, according to the first continuous scene image, inspection
The movement locus of measuring car；
Doubtful parking area detection module, if movement locus and a certain parking area for vehicle are overlapping, while vehicle into
The area for entering overlapping parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information output module, for the corresponding parking stall coordinate information of doubtful parking area to be sent to the second camera shooting
Machine.
Further, the vehicle ID and direction of motion acquisition module include：
Second continuous scene acquisition module is used for the parking stall coordinate information of the doubtful parking area according to reception, the second video camera
Automatically it furthers the region of corresponding parking stall coordinate information, obtains the second continuous scene image；
Vehicle feature recognition module, for other to the second continuous scene image progress Car license recognition and vehicle inscriptions, acquisition is per frame figure
The Car license recognition and vehicle money recognition result of picture；
Multiframe recognition result Fusion Module, for respectively melting the license plate recognition result of every frame image, vehicle money recognition result
It closes, obtains best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation modules, for other as a result, generating vehicle ID according to best license plate recognition result and optimized vehicle inscriptions；
Direction of vehicle movement detection module, for carrying out track of vehicle detection to vehicle ID in the second continuous scene image, if
The track of vehicle of vehicle ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if the vehicle of vehicle ID
Track is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
Further, the multiframe recognition result Fusion Module includes：For counting successively in the second continuous scene image
The license plate recognition result of each frame scene image, calculates the quantity of identical license plate recognition result, chooses identical license plate recognition result
License plate recognition result corresponding to quantity maximum value, as best license plate recognition result；The second continuous scene image is counted successively
In each frame scene image vehicle inscriptions not as a result, calculate the quantity of identical vehicle money recognition result, choose identical vehicle inscriptions and do not tie
Vehicle inscriptions corresponding to fruit quantity maximum value are not as a result, as optimized vehicle money recognition result.
Further, the vehicle ID generation modules are used for according to best license plate recognition result+optimized vehicle money recognition result
Form generate vehicle ID.
Further, the vehicle ID comparing modules include：
Car plate comparing module, the matching degree of the license plate recognition result for calculating the vehicle ID stored in vehicle ID and database,
If the matching degree of license plate recognition result is not less than third threshold value, then it is assumed that it compares successfully, is transferred to comparison result output module, it is no
Then it is transferred to vehicle money comparing module；
Vehicle money comparing module, the matching degree of the vehicle money recognition result for calculating the vehicle ID stored in vehicle ID and database,
If the matching degree of vehicle money recognition result is not less than the 4th threshold value, then it is assumed that it compares successfully, is transferred to comparison result output module, it is no
Then it is transferred to car plate vehicle money joint comparing module；
Car plate vehicle money combines comparing module, if the matching degree for license plate recognition result is not less than the 5th threshold value, and vehicle money
The matching degree of recognition result is not less than the 6th threshold value, then it is assumed that and it compares successfully, is transferred to comparison result output module, otherwise it is assumed that than
To failure；
Comparison result output module, if for comparing the relevant information for successfully recording vehicle ID and Parking；If than
To failure, then records vehicle ID and compare failure information.
Further, the third threshold value is more than the 5th threshold value, and the 4th threshold value is more than the 6th threshold value.
Compared with existing Roadside Parking detection technique, the invention has the beneficial effects that：Using the first video camera and
The combination of two video cameras can detect Parking and be captured to the vehicle of Parking；To the continuous vehicle figure of candid photograph
It is other as carrying out multiframe car plate and vehicle inscriptions, and multiframe recognition result is merged, license plate shading and nothing can be efficiently solved
Board vehicle problem.
Description of the drawings
Fig. 1 shows the flow chart of the Roadside Parking detection method according to the invention analyzed based on vehicle characteristics.
Fig. 2 shows the frame diagrams of the Roadside Parking detection device according to the invention based on vehicle characteristics analysis.
Specific implementation mode
To enable those skilled in the art to further appreciate that structure, feature and the other purposes of the present invention, in conjunction with institute
Detailed description are as follows for attached preferred embodiment, and illustrated preferred embodiment is only used to illustrate the technical scheme of the present invention, and is not limited
The fixed present invention.
Fig. 1 gives the flow chart of the Roadside Parking detection method according to the invention analyzed based on vehicle characteristics.Such as Fig. 1
Shown, the Roadside Parking detection method according to the invention based on vehicle characteristics analysis includes：
First step S1 obtains the first continuous scene image using the first video camera；
Second step S2 carries out doubtful parking area detection to the first continuous scene image, obtains the parking stall of doubtful parking area
Coordinate information, and it is sent to the second video camera；
Third step S3, the second video camera further parking stall coordinate information region automatically, the second continuous scene image are obtained, into driving
Board identifies and vehicle inscriptions are other, obtains best Car license recognition and vehicle inscriptions not as a result, generating vehicle ID, and carry out direction of motion inspection
It surveys；
Four steps S4, if the vehicle ID directions of motion are to drive into, by parking stall coordinate information, vehicle ID, direction of vehicle movement
It is sent to the first video camera, the second video camera is returned to original state, and is transferred to the 5th step S5；If the movement sides vehicle ID
To be driven out to, then the 6th step S6 is transferred to；
5th step S5 obtains the continuous scene image of third using the first video camera, vehicle ID track detections is carried out, if vehicle
The tracks ID do not update within the second threshold time, then store vehicle ID, parking stall coordinate information and down time and return, otherwise delete
Except vehicle ID and return；
6th step S6 obtains the 4th continuous scene image using the second video camera, carries out vehicle ID track detections, second is taken the photograph
Camera returns to original state, if the tracks vehicle ID are transferred to the 7th step S7, otherwise return far from parking stall coordinate；
Vehicle ID is compared 7th step S7 with the vehicle ID in database, if compared successfully, record vehicle ID and
The relevant information of Parking；If comparing failure, records vehicle ID and compare failure information.
Wherein, returned described in the 5th step S5 and the 6th step S6 refers to returning to the first step S1 weights
New start to process.
First video camera and the second video camera are mounted in trackside monitoring upright bar, and mounting height is located at ground 1.5 ~ 8
At rice, 2 ~ 12 parking stall ranges can be covered.Preferably, it is vertical to be mounted on trackside monitoring for first video camera and the second video camera
On bar, mounting height is located at 3 ~ 7 meters of ground, can cover 2 ~ 12 parking stall ranges.
Further, first video camera is the video camera that focal length is not less than 3.6mm, and second video camera is zoom
Video camera.Preferably, first video camera is the video camera that focal length is not less than 6mm.
Embodiment, it is the gun type camera of 8mm as the first video camera to use focal length, using ball-type video camera as second
Video camera is mounted in trackside monitoring upright bar, and mounting height is located at 5 meters of ground.
Further, the second step S2 includes：
Track of vehicle detecting step S21, using the method for moving object detection and tracking, according to the first continuous scene image, inspection
The movement locus of measuring car；
Doubtful parking area detecting step S22, if the movement locus of vehicle is overlapping with a certain parking area, while vehicle enters
The area of overlapping parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information exports step S23, and the corresponding parking stall coordinate information of doubtful parking area is sent to the second video camera.
Further, the track of vehicle detection can pass through existing track of vehicle detection method, device or equipment
It realizes.Embodiment, using " based on the matched video motion vehicle detection of adaptive profile and tracking Yang Jianguos, Yin Xuquan, side
It is beautiful, Li Jian, Wang Zhaoan《XI AN JIAOTONG UNIVERSITY Subject Index》, 2005, 39(4):Method in 351-355 " papers obtains vehicle
Movement locus.
Further, the value range of the first threshold is 0.1 ~ 4 square metre.Preferably, the first threshold takes
Ranging from 0.2 ~ 2 square metre of value.
Embodiment, the doubtful parking area detecting step S22 include：The movement locus of acquisition and vehicle occurs overlapping
Parking area, labeled as overlapping parking area；The area for calculating the vehicle sections in overlapping parking area, if in overlapping parking stall
The area of vehicle sections is more than 0.5 square metre in region, then it is doubtful parking area to mark and overlap parking area.
Further, the third step S3 includes：
Second continuous scene obtaining step S31, according to the parking stall coordinate information of the doubtful parking area of reception, the second video camera is certainly
The region of the dynamic corresponding parking stall coordinate information that furthers, obtains the second continuous scene image；
Vehicle feature recognition step S32 carries out Car license recognition to the second continuous scene image and vehicle inscriptions is other, obtains per frame image
Car license recognition and vehicle money recognition result；
Multiframe recognition result fusion steps S33 respectively melts the license plate recognition result of every frame image, vehicle money recognition result
It closes, obtains best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation step S34, according to best license plate recognition result and optimized vehicle inscriptions not as a result, generating vehicle ID；
Direction of vehicle movement detecting step S35 carries out track of vehicle detection, if vehicle to vehicle ID in the second continuous scene image
The track of vehicle of ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if the vehicle of vehicle ID
Track is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
Wherein, the Car license recognition in the vehicle feature recognition step S32 and vehicle inscriptions can not pass through prior art reality
It is existing.Embodiment, vehicle inscriptions Cai Yong not " the CN201610368032.6 a kind of recognition methods of vehicle money and dress based on convolutional neural networks
Set " in method realize.
Further, the multiframe recognition result fusion steps S33 includes：Count every in the second continuous scene image successively
The license plate recognition result of one frame scene image, calculates the quantity of identical license plate recognition result, chooses identical license plate recognition result number
The license plate recognition result corresponding to maximum value is measured, as best license plate recognition result；It counts in the second continuous scene image successively
The vehicle inscriptions of each frame scene image choose identical vehicle money recognition result not as a result, calculate the quantity of identical vehicle money recognition result
Vehicle inscriptions corresponding to quantity maximum value are not as a result, as optimized vehicle money recognition result.
Embodiment, the second continuous scene image have 10 frame images, and wherein license plate recognition result is " capital A12345 ", " capital
A12346 ", " capital A12356 " have 6 frames, 3 frames, 1 frame respectively, then choose Car license recognition " capital A12345 " and be used as best identified knot
Fruit；Wherein vehicle money recognition result is " white masses CC ", " white Honda Odyssey ", " white Audi A4 " have 4 frames, 3 respectively
Frame, 3 frames then choose vehicle inscriptions not " white masses CC " and are used as best identified result.
Further, the vehicle ID generation steps S34 is according to best license plate recognition result+optimized vehicle money recognition result
Form generates vehicle ID.Embodiment, best license plate recognition result are " capital A12345 ", and optimized vehicle money recognition result is that " white is big
Many CC ", then the vehicle ID generated are " capital A12345+ white masses CC ".
Embodiment, the direction of vehicle movement detecting step S35 is using " based on the matched video motion vehicle of adaptive profile
Detect and track Yang Jianguos, Yin Xuquan, Fang Li, Li Jian, Wang Zhaoan《XI AN JIAOTONG UNIVERSITY Subject Index》, 2005, 39(4):
Method in 351-355 " papers obtains the movement locus of vehicle ID, if the movement locus point of vehicle ID moves closer to parking stall
Coordinate, then it is assumed that the direction of motion of vehicle ID is to drive into；If the movement locus point of vehicle ID is gradually distance from parking stall coordinate, recognize
The direction of motion for vehicle ID is to be driven out to.
Further, the 5th step S5 includes：The continuous scene image of third is obtained using the first video camera；According to
Three continuous scene images carry out track of vehicle detection, if the track of vehicle of vehicle ID is in the time of second threshold to vehicle ID
It is not updated in section, then it is assumed that vehicle ID remains static, and vehicle ID, parking stall coordinate information and down time are stored in
In database and the first step S1 is returned, otherwise delete vehicle ID and returns to the first step S1.
Further, the value range of the second threshold is 3 ~ 60 seconds.Preferably, the value range of the second threshold
It is 5 ~ 20 seconds.
Further, the 6th step S6 includes：4th continuous scene image is obtained using the second video camera；According to
Four continuous scene images carry out track of vehicle detection to vehicle ID, the second video camera are returned to original state, if vehicle ID
Track of vehicle far from parking stall coordinate, then be transferred to the 7th step S7, otherwise return to the first step S1.
The 7th step S7 may be used existing comparison method and realize.
Further, the 7th step S7 includes：
Car plate compares step S71, calculates the matching degree of the license plate recognition result of the vehicle ID stored in vehicle ID and database, such as
The matching degree of fruit license plate recognition result is not less than third threshold value, then it is assumed that and it compares successfully, is transferred to comparison result output step S74,
Otherwise it is transferred to vehicle money and compares step S72；
Vehicle money compares step S72, calculates the matching degree of the vehicle money recognition result of the vehicle ID stored in vehicle ID and database, such as
The matching degree of fruit vehicle money recognition result is not less than the 4th threshold value, then it is assumed that and it compares successfully, is transferred to comparison result output step S74,
Otherwise car plate vehicle money Combined Ration is transferred to step S73；
Car plate vehicle money Combined Ration is to step S73, if the matching degree of license plate recognition result is not less than the 5th threshold value, and vehicle inscriptions
The matching degree of other result is not less than the 6th threshold value, then it is assumed that and it compares successfully, is transferred to comparison result output step S74, otherwise it is assumed that
Compare failure；
Comparison result exports step S74, if comparing the relevant information for successfully recording vehicle ID and Parking；If than
To failure, then records vehicle ID and compare failure information.
Further, the third threshold value is more than the 5th threshold value, and the 4th threshold value is more than the 6th threshold value.
The value range of the third threshold value is 0.7 ~ 0.95, and the value range of the 4th threshold value is 0.6 ~ 0.9, described
The value range of 5th threshold value is 0.5 ~ 0.7, and the value range of the 6th threshold value is 0.5 ~ 0.7.
Fig. 2 gives the frame diagram of the Roadside Parking detection device according to the invention analyzed based on vehicle characteristics.Such as Fig. 2
Shown, the Roadside Parking detection device according to the invention based on vehicle characteristics analysis includes：
First continuous scene image acquisition module 1, for obtaining the first continuous scene image using the first video camera；
Doubtful parking area parking stall coordinate obtaining module 2, for carrying out doubtful parking area detection to the first continuous scene image,
The parking stall coordinate information of doubtful parking area is obtained, and is sent to the second video camera；
Vehicle ID and direction of motion acquisition module 3 further parking stall coordinate information region automatically for the second video camera, obtain second
Continuous scene image, carries out Car license recognition and vehicle inscriptions are other, obtains best Car license recognition and vehicle inscriptions not as a result, generating vehicle
ID, and carry out direction of motion detection；
Selecting module 4 based on the direction of motion, if being to drive into for the vehicle ID directions of motion, by parking stall coordinate information, vehicle
ID, direction of vehicle movement are sent to the first video camera, and the second video camera is returned to original state, and are transferred to vehicle ID storages
With removing module 5；If the vehicle ID directions of motion are to be driven out to, it is transferred to vehicle and is driven out to processing module 6；
Vehicle ID storages and removing module 5 carry out vehicle ID rails for obtaining the continuous scene image of third using the first video camera
Mark detects, if the tracks vehicle ID do not update within the second threshold time, when storing vehicle ID, parking stall coordinate information and parking
Between and return, otherwise delete and vehicle ID and return；
Vehicle is driven out to processing module 6, for obtaining the 4th continuous scene image using the second video camera, carries out the inspection of the tracks vehicle ID
It surveys, the second video camera is returned to original state, if the tracks vehicle ID are transferred to vehicle ID comparing modules far from parking stall coordinate
7, otherwise return；
Vehicle ID comparing modules 7, if compared successfully, are remembered for vehicle ID to be compared with the vehicle ID in database
Record the relevant information of vehicle ID and Parking；If comparing failure, records vehicle ID and compare failure information.
Wherein, it refers to returning that the vehicle ID storages and removing module 5 and the vehicle, which are driven out to return described in processing module 6,
The first continuous scene image acquisition module 1 is returned to start the process over.
First video camera and the second video camera are mounted in trackside monitoring upright bar, and mounting height is located at ground 1.5 ~ 8
At rice, 2 ~ 12 parking stall ranges can be covered.
Further, first video camera is the video camera that focal length is not less than 3.6mm, and second video camera is zoom
Video camera.
Further, the doubtful parking area parking stall coordinate obtaining module 2 includes：
Track of vehicle detection module 21, for the method using moving object detection and tracking, according to the first continuous scene image,
Detect the movement locus of vehicle；
Doubtful parking area detection module 22, if it is overlapping for the movement locus of vehicle and a certain parking area, while vehicle
Area into overlapping parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information output module 23, for the corresponding parking stall coordinate information of doubtful parking area to be sent to the second camera shooting
Machine.
Further, the value range of the first threshold is 0.1 ~ 4 square metre.Preferably, the first threshold takes
Ranging from 0.2 ~ 2 square metre of value.
Further, the vehicle ID and direction of motion acquisition module 3 include：
Second continuous scene acquisition module 31 is used for the parking stall coordinate information of the doubtful parking area according to reception, the second camera shooting
Machine furthers the region of corresponding parking stall coordinate information automatically, obtains the second continuous scene image；
Vehicle feature recognition module 32, for other to the second continuous scene image progress Car license recognition and vehicle inscriptions, acquisition is per frame
The Car license recognition and vehicle money recognition result of image；
Multiframe recognition result Fusion Module 33, for respectively carrying out the license plate recognition result of every frame image, vehicle money recognition result
Fusion, obtains best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation modules 34, for other as a result, generating vehicle ID according to best license plate recognition result and optimized vehicle inscriptions；
Direction of vehicle movement detection module 35, for carrying out track of vehicle detection to vehicle ID in the second continuous scene image, such as
The track of vehicle of fruit vehicle ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if vehicle ID
Track of vehicle is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
Further, the multiframe recognition result Fusion Module 33 includes：For counting the second continuous scene image successively
In each frame scene image license plate recognition result, calculate the quantity of identical license plate recognition result, choose identical Car license recognition knot
License plate recognition result corresponding to fruit quantity maximum value, as best license plate recognition result；The second continuous scene graph is counted successively
It is other to choose identical vehicle inscriptions not as a result, calculate the quantity of identical vehicle money recognition result for the vehicle inscriptions of each frame scene image as in
Vehicle inscriptions corresponding to fruiting quantities maximum value are not as a result, as optimized vehicle money recognition result.
Further, the vehicle ID generation modules 34 according to best license plate recognition result+optimized vehicle inscriptions for not tying
The form of fruit generates vehicle ID.
Further, the value range of the second threshold is 3 ~ 60 seconds.Preferably, the value range of the second threshold
It is 5 ~ 20 seconds.
Further, the vehicle ID comparing modules 7 include：
Car plate comparing module 71, the matching of the license plate recognition result for calculating the vehicle ID stored in vehicle ID and database
Degree, if the matching degree of license plate recognition result is not less than third threshold value, then it is assumed that compare successfully, be transferred to comparison result output module
74, otherwise it is transferred to vehicle money comparing module 72；
Vehicle money comparing module 72, the matching of the vehicle money recognition result for calculating the vehicle ID stored in vehicle ID and database
Degree, if the matching degree of vehicle money recognition result is not less than the 4th threshold value, then it is assumed that compare successfully, be transferred to comparison result output module
74, otherwise it is transferred to car plate vehicle money joint comparing module 73；
Car plate vehicle money combines comparing module 73, if the matching degree for license plate recognition result is not less than the 5th threshold value, and vehicle
The matching degree of money recognition result is not less than the 6th threshold value, then it is assumed that compares successfully, is transferred to comparison result output module 74, otherwise recognizes
Fail to compare；
Comparison result output module 74, if for comparing the relevant information for successfully recording vehicle ID and Parking；If
Failure is compared, then record vehicle ID and compares failure information.
Further, the third threshold value is more than the 5th threshold value, and the 4th threshold value is more than the 6th threshold value.
The value range of the third threshold value is 0.7 ~ 0.95, and the value range of the 4th threshold value is 0.6 ~ 0.9, described
The value range of 5th threshold value is 0.5 ~ 0.7, and the value range of the 6th threshold value is 0.5 ~ 0.7.
Compared with existing Roadside Parking detection technique, the invention has the beneficial effects that：Using the first video camera and
The combination of two video cameras can detect Parking and be captured to the vehicle of Parking；To the continuous vehicle figure of candid photograph
It is other as carrying out multiframe car plate and vehicle inscriptions, and multiframe recognition result is merged, license plate shading and nothing can be efficiently solved
Board vehicle problem.
The foregoing is only a preferred embodiment of the present invention, is not intended to limit the scope of the present invention, should
Understand, the present invention is not limited to implementation as described herein, the purpose of these implementations description is to help this field
In technical staff put into practice the present invention.Any those of skill in the art are easy to do not departing from spirit and scope of the invention
In the case of be further improved and perfect, therefore the present invention is only by the content of the claims in the present invention and limiting for range
System, intention, which covers, all to be included the alternative in the spirit and scope of the invention being defined by the appended claims and waits
Same scheme.
Claims (12)
1. the Roadside Parking detection method based on vehicle characteristics analysis, which is characterized in that this method includes：
First step obtains the first continuous scene image using the first video camera；
Second step carries out doubtful parking area detection to the first continuous scene image, and the parking stall for obtaining doubtful parking area is sat
Information is marked, and is sent to the second video camera；
Third step, the second video camera further parking stall coordinate information region automatically, obtain the second continuous scene image, carry out car plate
Identification and vehicle inscriptions are other, obtain best Car license recognition and vehicle inscriptions not as a result, generating vehicle ID, and carry out direction of motion detection；
Four steps sends out parking stall coordinate information, vehicle ID, direction of vehicle movement if the vehicle ID directions of motion are to drive into
The first video camera is given, the second video camera is returned to original state, and is transferred to the 5th step；If the vehicle ID directions of motion are
It is driven out to, is then transferred to the 6th step；
5th step obtains the continuous scene image of third using the first video camera, vehicle ID track detections is carried out, if vehicle ID
Track does not update within the second threshold time, then stores vehicle ID, parking stall coordinate information and down time and return, otherwise delete
Vehicle ID is simultaneously returned；
6th step obtains the 4th continuous scene image using the second video camera, carries out vehicle ID track detections, by the second camera shooting
Machine returns to original state, if the tracks vehicle ID are transferred to the 7th step, otherwise return far from parking stall coordinate；
Vehicle ID is compared 7th step with the vehicle ID in database, if compared successfully, records vehicle ID and stops
The relevant information of vehicle event；If comparing failure, records vehicle ID and compare failure information；
Wherein, the return refers to returning to first step.
2. the method as described in claim 1, which is characterized in that first video camera and the second video camera are supervised mounted on trackside
It controls in upright bar, mounting height is located at 1.5 ~ 8 meters of ground, can cover 2 ~ 12 parking stall ranges；Further, it described first takes the photograph
Camera is the video camera that focal length is not less than 3.6mm, and second video camera is Zoom camera.
3. the method as described in claim 1, which is characterized in that the second step includes：
Track of vehicle detecting step detects vehicle using the method for moving object detection and tracking according to the first continuous scene image
Movement locus；
Doubtful parking area detecting step, if the movement locus of vehicle is overlapping with a certain parking area, while vehicle enters friendship
The area of folded parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information exports step, and the corresponding parking stall coordinate information of doubtful parking area is sent to the second video camera.
4. the method as described in claim 1, which is characterized in that the third step includes：
Second continuous scene obtaining step, according to the parking stall coordinate information of the doubtful parking area of reception, the second video camera is automatic
It furthers the region of corresponding parking stall coordinate information, obtains the second continuous scene image；
Vehicle feature recognition step carries out Car license recognition to the second continuous scene image and vehicle inscriptions is other, obtains per frame image
Car license recognition and vehicle money recognition result；
The license plate recognition result of every frame image, vehicle money recognition result are merged, are obtained by multiframe recognition result fusion steps respectively
Take best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation steps, according to best license plate recognition result and optimized vehicle inscriptions not as a result, generating vehicle ID；
Direction of vehicle movement detecting step carries out track of vehicle detection, if vehicle to vehicle ID in the second continuous scene image
The track of vehicle of ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if the vehicle rail of vehicle ID
Mark is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
5. method as claimed in claim 4, which is characterized in that the multiframe recognition result fusion steps include：It counts successively
The license plate recognition result of each frame scene image in second continuous scene image, calculates the quantity of identical license plate recognition result, choosing
The license plate recognition result corresponding to identical license plate recognition result quantity maximum value is taken, as best license plate recognition result；It unites successively
The vehicle inscriptions of each frame scene image in the second continuous scene image are counted not as a result, calculate the quantity of identical vehicle money recognition result,
The vehicle inscriptions corresponding to the identical other fruiting quantities maximum value of vehicle inscriptions are chosen not as a result, as optimized vehicle money recognition result.
6. method as claimed in claim 4, which is characterized in that the vehicle ID generation steps are according to best license plate recognition result
The form of+optimized vehicle money recognition result generates vehicle ID.
7. the method as described in claim 1, which is characterized in that the 7th step includes：
Car plate compares step, calculates the matching degree of the license plate recognition result of the vehicle ID stored in vehicle ID and database, if
The matching degree of license plate recognition result is not less than third threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, otherwise turns
Enter vehicle money and compares step；
Vehicle money compares step, calculates the matching degree of the vehicle money recognition result of the vehicle ID stored in vehicle ID and database, if
The matching degree of vehicle money recognition result is not less than the 4th threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, otherwise turns
Enter car plate vehicle money Combined Ration to step；
Car plate vehicle money Combined Ration is to step, if the matching degree of license plate recognition result is not less than the 5th threshold value, and vehicle inscriptions are other
As a result matching degree is not less than the 6th threshold value, then it is assumed that compares successfully, is transferred to comparison result output step, is lost otherwise it is assumed that comparing
It loses；
Comparison result exports step, if comparing the relevant information for successfully recording vehicle ID and Parking；It is lost if compared
It loses, then record vehicle ID and compares failure information；
Wherein, the third threshold value is more than the 5th threshold value, and the 4th threshold value is more than the 6th threshold value.
8. the value range of the method as described in claim 1 ~ 7, the first threshold is 0.1 ~ 4 square metre, second threshold
The value range of value is 3 ~ 60 seconds, and the value range of the third threshold value is 0.7 ~ 0.95, the value range of the 4th threshold value
It is 0.6 ~ 0.9, the value range of the 5th threshold value is 0.5 ~ 0.7, and the value range of the 6th threshold value is 0.5 ~ 0.7.
9. the Roadside Parking detection device based on vehicle characteristics analysis, which is characterized in that the device includes：
First continuous scene image acquisition module, for obtaining the first continuous scene image using the first video camera；
Doubtful parking area parking stall coordinate obtaining module, for carrying out doubtful parking area detection to the first continuous scene image,
The parking stall coordinate information of doubtful parking area is obtained, and is sent to the second video camera；
Vehicle ID and direction of motion acquisition module further parking stall coordinate information region automatically for the second video camera, obtain second
Continuous scene image, carries out Car license recognition and vehicle inscriptions are other, obtains best Car license recognition and vehicle inscriptions not as a result, generating vehicle
ID, and carry out direction of motion detection；
Selecting module based on the direction of motion, if being to drive into for the vehicle ID directions of motion, by parking stall coordinate information, vehicle
ID, direction of vehicle movement are sent to the first video camera, and the second video camera is returned to original state, and be transferred to vehicle ID storage and
Removing module；If the vehicle ID directions of motion are to be driven out to, it is transferred to vehicle and is driven out to processing module；
Vehicle ID storages and removing module carry out vehicle ID rails for obtaining the continuous scene image of third using the first video camera
Mark detects, if the tracks vehicle ID do not update within the second threshold time, when storing vehicle ID, parking stall coordinate information and parking
Between and return, otherwise delete and vehicle ID and return；
Vehicle is driven out to processing module, for obtaining the 4th continuous scene image using the second video camera, carries out the inspection of the tracks vehicle ID
It surveys, the second video camera is returned to original state, if the tracks vehicle ID are transferred to vehicle ID and compare mould far from parking stall coordinate
Otherwise block returns；
Vehicle ID comparing modules, if compared successfully, are recorded for vehicle ID to be compared with the vehicle ID in database
The relevant information of vehicle ID and Parking；If comparing failure, records vehicle ID and compare failure information.
10. device as claimed in claim 9, which is characterized in that the doubtful parking area parking stall coordinate obtaining module includes：
Track of vehicle detection module, for the method using moving object detection and tracking, according to the first continuous scene image, inspection
The movement locus of measuring car；
Doubtful parking area detection module, if movement locus and a certain parking area for vehicle are overlapping, while vehicle into
The area for entering overlapping parking area is more than first threshold, then it is doubtful parking area to mark and overlap parking area；
Parking stall coordinate information output module, for the corresponding parking stall coordinate information of doubtful parking area to be sent to the second camera shooting
Machine.
11. device as claimed in claim 9, which is characterized in that the vehicle ID and direction of motion acquisition module include：
Second continuous scene acquisition module is used for the parking stall coordinate information of the doubtful parking area according to reception, the second video camera
Automatically it furthers the region of corresponding parking stall coordinate information, obtains the second continuous scene image；
Vehicle feature recognition module, for other to the second continuous scene image progress Car license recognition and vehicle inscriptions, acquisition is per frame figure
The Car license recognition and vehicle money recognition result of picture；
Multiframe recognition result Fusion Module, for respectively melting the license plate recognition result of every frame image, vehicle money recognition result
It closes, obtains best license plate recognition result and optimized vehicle money recognition result；
Vehicle ID generation modules, for other as a result, generating vehicle ID according to best license plate recognition result and optimized vehicle inscriptions；
Direction of vehicle movement detection module, for carrying out track of vehicle detection to vehicle ID in the second continuous scene image, if
The track of vehicle of vehicle ID is towards doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to drive into, if the vehicle of vehicle ID
Track is far from doubtful parking area, then it is assumed that the direction of motion of vehicle ID is to be driven out to.
12. device as claimed in claim 9, which is characterized in that the vehicle ID comparing modules include：
Car plate comparing module, the matching degree of the license plate recognition result for calculating the vehicle ID stored in vehicle ID and database,
If the matching degree of license plate recognition result is not less than third threshold value, then it is assumed that it compares successfully, is transferred to comparison result output module, it is no
Then it is transferred to vehicle money comparing module；
Vehicle money comparing module, the matching degree of the vehicle money recognition result for calculating the vehicle ID stored in vehicle ID and database,
If the matching degree of vehicle money recognition result is not less than the 4th threshold value, then it is assumed that it compares successfully, is transferred to comparison result output module, it is no
Then it is transferred to car plate vehicle money joint comparing module；
Car plate vehicle money combines comparing module, if the matching degree for license plate recognition result is not less than the 5th threshold value, and vehicle money
The matching degree of recognition result is not less than the 6th threshold value, then it is assumed that and it compares successfully, is transferred to comparison result output module, otherwise it is assumed that than
To failure；
Comparison result output module, if for comparing the relevant information for successfully recording vehicle ID and Parking；If than
To failure, then records vehicle ID and compare failure information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810203056.5A CN108305466B (en) | 2018-03-13 | 2018-03-13 | Roadside parking detection method and device based on vehicle characteristic analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810203056.5A CN108305466B (en) | 2018-03-13 | 2018-03-13 | Roadside parking detection method and device based on vehicle characteristic analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108305466A true CN108305466A (en) | 2018-07-20 |
| CN108305466B CN108305466B (en) | 2020-05-08 |
Family
ID=62849712
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810203056.5A Active CN108305466B (en) | 2018-03-13 | 2018-03-13 | Roadside parking detection method and device based on vehicle characteristic analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108305466B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109147341A (en) * | 2018-09-14 | 2019-01-04 | 杭州数梦工场科技有限公司 | Violation vehicle detection method and device |
| CN109523642A (en) * | 2018-09-17 | 2019-03-26 | 上海航天设备制造总厂有限公司 | A kind of curb parking intelligent toll system based on license plate recognition technology |
| CN109615858A (en) * | 2018-12-21 | 2019-04-12 | 深圳信路通智能技术有限公司 | A kind of intelligent parking behavior judgment method based on deep learning |
| CN110956644A (en) * | 2018-09-27 | 2020-04-03 | 杭州海康威视数字技术股份有限公司 | Motion trail determination method and system |
| CN111340710A (en) * | 2019-12-31 | 2020-06-26 | 智慧互通科技有限公司 | Method and system for acquiring vehicle information based on image stitching |
| CN111461124A (en) * | 2020-03-02 | 2020-07-28 | 浙江省北大信息技术高等研究院 | Large data-based shielded license plate recognition method and device and storage medium |
| CN111523385A (en) * | 2020-03-20 | 2020-08-11 | 北京航空航天大学合肥创新研究院 | Stationary vehicle detection method and system based on frame difference method |
| CN111626225A (en) * | 2020-05-28 | 2020-09-04 | 济南博观智能科技有限公司 | License plate recognition method, device and equipment for station vehicle and storage medium |
| CN113076797A (en) * | 2021-02-24 | 2021-07-06 | 江苏濠汉信息技术有限公司 | Charging station electric vehicle fire alarm method and system based on intelligent video identification |
| CN113450575A (en) * | 2021-05-31 | 2021-09-28 | 超级视线科技有限公司 | Management method and device for roadside parking |
| CN113570872A (en) * | 2021-08-13 | 2021-10-29 | 深圳市捷顺科技实业股份有限公司 | Processing method and device for blocking parking space event |
| CN114170810A (en) * | 2021-12-28 | 2022-03-11 | 深圳市捷顺科技实业股份有限公司 | Vehicle traveling direction identification method, system and device |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5343237A (en) * | 1990-10-08 | 1994-08-30 | Matsushita Electric Industrial Co., Ltd. | System for detecting and warning an illegally parked vehicle |
| JP2003058925A (en) * | 2001-08-17 | 2003-02-28 | Toshiba Corp | Parking lot management system |
| US7026954B2 (en) * | 2003-06-10 | 2006-04-11 | Bellsouth Intellectual Property Corporation | Automated parking director systems and related methods |
| CN101206799A (en) * | 2006-12-20 | 2008-06-25 | 索尼株式会社 | Monitoring system, monitoring apparatus and monitoring method |
| US20090157260A1 (en) * | 2007-12-12 | 2009-06-18 | Hyundai Motor Company | Automatic parking system for vehicle |
| CN202584422U (en) * | 2012-03-23 | 2012-12-05 | 罗普特（厦门）科技集团有限公司 | Apparatus for snapshotting vehicles violating traffic rules based on intelligent analysis and video monitoring |
| CN103473926A (en) * | 2013-09-11 | 2013-12-25 | 无锡加视诚智能科技有限公司 | Gun-ball linkage road traffic parameter collection and rule breaking snapshooting system |
| US20140236786A1 (en) * | 2012-08-06 | 2014-08-21 | Cloudparc, Inc. | Human Review of an Image Stream for a Parking Camera System |
| CN105809972A (en) * | 2016-03-24 | 2016-07-27 | 牛力伟 | Parking management method, device and system |
| CN106952477A (en) * | 2017-04-26 | 2017-07-14 | 智慧互通科技有限公司 | Roadside Parking management method based on polyphaser image Combined Treatment |
| CN107274677A (en) * | 2017-08-03 | 2017-10-20 | 智慧互通科技有限公司 | The Roadside Parking management system and method linked based on bar interdigit rifle ball video camera |
-
2018
- 2018-03-13 CN CN201810203056.5A patent/CN108305466B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5343237A (en) * | 1990-10-08 | 1994-08-30 | Matsushita Electric Industrial Co., Ltd. | System for detecting and warning an illegally parked vehicle |
| JP2003058925A (en) * | 2001-08-17 | 2003-02-28 | Toshiba Corp | Parking lot management system |
| US7026954B2 (en) * | 2003-06-10 | 2006-04-11 | Bellsouth Intellectual Property Corporation | Automated parking director systems and related methods |
| CN101206799A (en) * | 2006-12-20 | 2008-06-25 | 索尼株式会社 | Monitoring system, monitoring apparatus and monitoring method |
| US20090157260A1 (en) * | 2007-12-12 | 2009-06-18 | Hyundai Motor Company | Automatic parking system for vehicle |
| CN202584422U (en) * | 2012-03-23 | 2012-12-05 | 罗普特（厦门）科技集团有限公司 | Apparatus for snapshotting vehicles violating traffic rules based on intelligent analysis and video monitoring |
| US20140236786A1 (en) * | 2012-08-06 | 2014-08-21 | Cloudparc, Inc. | Human Review of an Image Stream for a Parking Camera System |
| CN103473926A (en) * | 2013-09-11 | 2013-12-25 | 无锡加视诚智能科技有限公司 | Gun-ball linkage road traffic parameter collection and rule breaking snapshooting system |
| CN105809972A (en) * | 2016-03-24 | 2016-07-27 | 牛力伟 | Parking management method, device and system |
| CN106952477A (en) * | 2017-04-26 | 2017-07-14 | 智慧互通科技有限公司 | Roadside Parking management method based on polyphaser image Combined Treatment |
| CN107274677A (en) * | 2017-08-03 | 2017-10-20 | 智慧互通科技有限公司 | The Roadside Parking management system and method linked based on bar interdigit rifle ball video camera |
Non-Patent Citations (2)
| Title |
|---|
| 司博章 等: "一种路侧停车位物联网管理***", 《物联网技术》 * |
| 茅嘉磊: "基于视频分析的路侧停车自动计时取证", 《上海船舶运输科学研究所学报》 * |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109147341B (en) * | 2018-09-14 | 2019-11-22 | 杭州数梦工场科技有限公司 | Violation vehicle detection method and device |
| CN109147341A (en) * | 2018-09-14 | 2019-01-04 | 杭州数梦工场科技有限公司 | Violation vehicle detection method and device |
| CN109523642A (en) * | 2018-09-17 | 2019-03-26 | 上海航天设备制造总厂有限公司 | A kind of curb parking intelligent toll system based on license plate recognition technology |
| CN110956644B (en) * | 2018-09-27 | 2023-10-10 | 杭州海康威视数字技术股份有限公司 | Motion trail determination method and system |
| CN110956644A (en) * | 2018-09-27 | 2020-04-03 | 杭州海康威视数字技术股份有限公司 | Motion trail determination method and system |
| CN109615858A (en) * | 2018-12-21 | 2019-04-12 | 深圳信路通智能技术有限公司 | A kind of intelligent parking behavior judgment method based on deep learning |
| CN111340710A (en) * | 2019-12-31 | 2020-06-26 | 智慧互通科技有限公司 | Method and system for acquiring vehicle information based on image stitching |
| CN111340710B (en) * | 2019-12-31 | 2023-11-07 | 智慧互通科技股份有限公司 | Method and system for acquiring vehicle information based on image stitching |
| CN111461124A (en) * | 2020-03-02 | 2020-07-28 | 浙江省北大信息技术高等研究院 | Large data-based shielded license plate recognition method and device and storage medium |
| CN111523385A (en) * | 2020-03-20 | 2020-08-11 | 北京航空航天大学合肥创新研究院 | Stationary vehicle detection method and system based on frame difference method |
| CN111626225A (en) * | 2020-05-28 | 2020-09-04 | 济南博观智能科技有限公司 | License plate recognition method, device and equipment for station vehicle and storage medium |
| CN111626225B (en) * | 2020-05-28 | 2022-05-17 | 济南博观智能科技有限公司 | License plate recognition method, device and equipment for station vehicle and storage medium |
| CN113076797A (en) * | 2021-02-24 | 2021-07-06 | 江苏濠汉信息技术有限公司 | Charging station electric vehicle fire alarm method and system based on intelligent video identification |
| CN113450575A (en) * | 2021-05-31 | 2021-09-28 | 超级视线科技有限公司 | Management method and device for roadside parking |
| CN113570872A (en) * | 2021-08-13 | 2021-10-29 | 深圳市捷顺科技实业股份有限公司 | Processing method and device for blocking parking space event |
| CN114170810A (en) * | 2021-12-28 | 2022-03-11 | 深圳市捷顺科技实业股份有限公司 | Vehicle traveling direction identification method, system and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108305466B (en) | 2020-05-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108305466A (en) | Roadside Parking detection method and device based on vehicle characteristics analysis | |
| CN106952477B (en) | Roadside parking management method based on multi-camera image joint processing | |
| CN105069429B (en) | A kind of flow of the people analytic statistics methods and system based on big data platform | |
| CN108491758B (en) | Track detection method and robot | |
| CN104574954B (en) | A kind of vehicle auditing method, control device and system based on free streaming system | |
| CN105139425B (en) | A kind of demographic method and device | |
| US8098290B2 (en) | Multiple camera system for obtaining high resolution images of objects | |
| CN107945521A (en) | A kind of rifle ball linkage trackside is parked detecting system and method | |
| CN107316462A (en) | A kind of flow statistical method and device | |
| CN108198430A (en) | A kind of trackside fence parking management system based on video camera and sensor | |
| CN101789177B (en) | Device and method for detecting and tracking vehicles crossing and pressing the yellow line and for capturing vehicle information | |
| US20200349348A1 (en) | Method for person re-identification in enclosed place, system, and terminal device | |
| CN105913367A (en) | Public bus passenger flow volume detection system and method based on face identification and position positioning | |
| US20140204206A1 (en) | Line scan imaging from a raw video source | |
| CN105491327A (en) | Video tracking method and device based on road network | |
| CN112084928B (en) | Road traffic accident detection method based on visual attention mechanism and ConvLSTM network | |
| CN107480653A (en) | passenger flow volume detection method based on computer vision | |
| CN102426785A (en) | Traffic flow information perception method based on contour and local characteristic point and system thereof | |
| CN108091163A (en) | A kind of Roadside Parking management system combined based on video camera with trackside fence | |
| CN116245911B (en) | Video offline statistics method | |
| CN107316463A (en) | A kind of method and apparatus of vehicle monitoring | |
| CN110460813A (en) | A kind of container representation acquisition device and acquisition method based on video flowing | |
| CN108235017A (en) | For the method and apparatus of detecting event | |
| CN208092975U (en) | The system for realizing semiclosed parking lot management based on long short focus camera | |
| CN102565103A (en) | Tracking detection method for weld defects based on X-ray image |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |