CN201345166Y - Video-based passage vehicle detector - Google Patents
Video-based passage vehicle detector Download PDFInfo
- Publication number
- CN201345166Y CN201345166Y CNU2008202338582U CN200820233858U CN201345166Y CN 201345166 Y CN201345166 Y CN 201345166Y CN U2008202338582 U CNU2008202338582 U CN U2008202338582U CN 200820233858 U CN200820233858 U CN 200820233858U CN 201345166 Y CN201345166 Y CN 201345166Y
- Authority
- CN
- China
- Prior art keywords
- video
- binocular
- processor
- passage
- memory
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The utility model relates to a video-based passage vehicle detector, belonging to the technical field of urban indoor parking lot passage vehicle detection. The detector consists of a binocular video capture device, a video memory, a video processor and a network interface. The binocular video capture device includes two sets of cameras and an A/D converting circuit. The two cameras are respectively used to pick up the images in the corresponding passage and the view field of one camera is overlapped with the view field of the other camera to some extent. The digital signal output interface of the A/D converting circuit is connected with the video digital input interface of the video memory. The video digital output interface of the video memory is connected with the video processor. The video memory transmits the received passage video digital image signal to the video processor for analysis, so as to extract relevant information. The output port of the video processor is connected with a host computer through the network interface. The utility model effectively resolves the problem of the visual angle dead zone of the central part between the two passages, and has the advantages of low cost, high practicability, high accuracy, simple installation and maintenance and the like.
Description
Technical field
The video channel wagon detector belongs to passage vehicle detection technique field, parking garage, city.
Background technology
Usually on duplex channel, the vehicle that standard is driven only can travel in some passages, and therefore, what conventional passage wagon detector adopted is the single channel detection mode, and each detecting device only detects the vehicle on this passage.And the vehicle ' situation more complicated of parking garage, the parking garage is adopted duplex channel usually, and interchannel do not have isolation guardrail, because the driving behavior of driver in the parking lot is lack of standardization, so vehicle travels between two passages through regular meeting.In this case, conventional passage vehicle detection mode obviously can not satisfy the requirement of vehicle detection, the situation of a large amount of omissions, flase drop can occur.
Traditional single channel vehicle checking method commonly used mainly contains following two kinds:
(1) inductive coil detects: when the vehicle admission passage, be installed on underground coil and produce electromagnetic induction signal, testing circuit is judged the situation of vehicle access way according to the time of detection signal generation and disappearance.This method working stability, the accuracy of detection height, but road pavement has damage, construction and installation inconvenience, and poor expandability can not regulate and move, and maintenance difficulties is big.
(2) infrared detection: transmit and receive the synchronous scanning that realizes vehicle by linearly aligned infrared light, and light signal is converted to electric signal, thereby realization is to the comprehensive detection of vehicle data, infra red vehicle detector product technology maturation, simple installation, high-speed response, strong interference immunity.But because detection sensitivity is relevant with thermal emissivity rate, infrared detection is subjected to the influence of the thermal source of vehicle own, and antimierophonic ability is poor, and is strict to time-temperature relation when detecting.
The utility model content
The purpose of this utility model is to overcome the above-mentioned defective of existing wagon detector, a kind of passage wagon detector based on video is provided, this detecting device has solved the visual angle blind zone problem of binary channels central authorities, also has advantages such as installation is simple, maintenance cost is low simultaneously.
To achieve these goals, the utility model has been taked following technical scheme.The video channel wagon detector utilizes the composograph of two video cameras to carry out vehicle target identification and detects, and specifically comprises binocular video collector, video memory, video processor and the network interface that is connected with host computer;
Described binocular video collector, by two video cameras and A the D change-over circuit constitute.Distance between two video cameras is 15 centimetres～30 centimetres; Two video cameras are taken the image in its respective channel respectively, and have certain coincidence visual field, and the width range that overlaps the visual field is 15 centimetres～30 centimetres.Two simulating signals that video camera absorbed change into data image signal by A D change-over circuit.
A the digital signal output interface of D change-over circuit link to each other with the video data input interface of video memory, the passage video digital images information that collects is exported to video memory.
The video data output interface of video memory links to each other with video processor, send the passage video digital images signal that receives to video processor, video processor is done analyzing and processing to the binocular video image that receives and is extracted relevant information, be specially: the recognizer of utilization moving object is earlier handled, judge time and the travel direction of vehicle into and out of sensing range, two picture frames that video camera absorbed splice when utilizing the image hybrid process that vehicle is in the optimum detection position again, extract license board information with the car plate recognizer at last.
The output port of video processor links to each other with host computer by network interface, the various information that video processor is extracted sends to host computer by network interface, and receive the control life that host computer sends, binocular video collector, video memory are controlled.
Passage wagon detector in the utility model has following advantage:
1) adopts video image processing technology.Compare with the conventional channels vehicle checking method, have many good qualities based on the detection technique of video: it belongs to the face formula detects, and sensing range is big; Installation, easy construction need not to destroy ground, and maintenance is simple, extendability is strong, precision is high, and visible image information can be provided;
2) have the car plate recognition function, thereby can realize the tracking of vehicle driving trace in the parking lot and induce, this is that the conventional sense device is not available.
3) solved the visual angle blind zone problem of binary channels central authorities.In the parking lot, because the complicacy of driver's driving behavior, vehicle often is not to be fixed in certain passage to travel, and often travel on link positions in the middle of two passages, if the incomplete situation of information of vehicles in the video information of shot by camera promptly can occur according to camera supervised passage of conventional method, thereby can cause follow-up vehicle and car plate identification to detect failure.The utility model utilizes the binocular video collector, monitor two passages simultaneously with two video cameras, make it have certain coincidence visual field, again two picture frames that video camera is taken the photograph are spliced the image that obtains containing complete license board information and carry out information of vehicles and extract, effectively solved the central visual angle of the binary channels blind zone problem parking lot in.
4) management maintenance is convenient and expense is low, and according to measured result, it is 99.8% that passage vehicle detection accuracy rate of the present utility model can reach in the parking garage.
Description of drawings
Fig. 1 video frequency vehicle detection system block diagram (being the utility model system architecture diagram in the frame of broken lines);
Fig. 2 binocular video collector interior video cameras level inclination synoptic diagram;
Fig. 3 binocular video collector interior video cameras vertical dip angle synoptic diagram;
Fig. 4 virtual detection line location drawing;
The process flow diagram of Fig. 5 vehicle detection;
Fig. 6 sets up flow process based on the binocular two field picture split-join model of Feature Points Matching.
Embodiment
The utility model is described in further detail below in conjunction with Fig. 1～Fig. 6:
As shown in Figure 1, present embodiment comprises binocular video collector, video memory, video processor and network interface.
The binocular video collector, by two video cameras and A the D change-over circuit constitute.Support with 20 centimetres between the video camera connects; The coincidence visual field of two video cameras is 30 centimetres; Two video cameras are taken vehicle discrepancy video image in the passage simultaneously.Close under the command condition of detecting device not receiving host computer, video collector will be monitored affiliated passage always.
The level inclination of two video cameras overlaps the visual field with it width concerns as shown in Figure 2.H1 overlaps the visual field width for the binocular video collector, and A is the X deflection angle of video camera, and L is the distance of subject to camera lens, and C1 is a video camera 1, and C2 is a video camera 2, and h2 is two distances between the video camera.
The vertical dip angle of two video cameras and the height relationships of its shot object are as shown in Figure 3.Among the figure: H is the height of subject, and L is the distance of subject to camera lens, H1 be video camera in the binocular video collector apart from the vertical height on ground, A is the vertical deflection angle of video camera, Vi is a video camera.
Video A the D converter adopt SAA7113.Have two analog video input ports, be used for connecting two simulation ccd video cameras; Be provided with the digital signal output interface, convert the analog video signal of gathering to digital video image and output to the video memory preservation, wait for next step processing.
The video image storer is made of storer (61LV51216) and control circuit (XC2S200).The digital video input interface of video memory is connected with the binocular video collector, receives the data image signal that comprises the passage video information; The digital video output interface is connected with video processor, sends the data image signal that receives to video processor and handles.
Video image processor by CPU (AT912M9200), data-carrier store (57V281620HC) and program storage (29LV320d) and corresponding I the O interface constitute.The digital picture that video image processor utilizes the binocular video collector to collect is synthesized, and obtains comprising the binary channels image of complete vehicle and license board information, carries out vehicle identification and license board information again and extracts.The entire process process comprises that mainly moving object identification and travel direction thereof are judged, the binocular picture frame splices and vehicle license plate characteristic extracts three parts:
The judgement of moving object identification and travel direction thereof such as Fig. 4, shown in Figure 5 at first handle the image utilization moving object recognizer of gathering.Three virtual detection lines are set earlier, utilize the grey scale change on three virtual detection lines of background frame difference method detection, if transformed value then thinks have vehicle to enter surveyed area greater than a certain threshold value, judge the travel direction of vehicle again by the time sequencing of three detection lines according to vehicle, if vehicle earlier by detection line 1 again by detection line 2 then think that vehicle is forward to travel, if vehicle passing detection line 3 is again by detection line 2 then think that vehicle is a reverse driving.Idiographic flow is as shown in Figure 5: in that car mark is arranged is fictitious time, whether the result that recycle ratio is carried out the background subtraction computing than the gray-scale value and the background template of 1 position of detection line in the current frame image and detection line 3 positions is greater than a threshold value (detection line 1 relatively earlier, detection line 1 comparative result less than the situation of threshold value under detection line 3 relatively again), if the result is greater than threshold value then think have car to enter detection zone, to there be car mark to be made as very, and record current time and vehicle heading.If detection line 1 position, shows that vehicle for forward travelling, if make detection line 3 positions greater than threshold value, shows that vehicle is a reverse driving greater than threshold value.In that car mark is arranged is true time, whether the result that recycle ratio is carried out the background subtraction computing than the gray-scale value and the background template of the pairing image-region in 2 positions of detection line in the current frame image is greater than a threshold value, if the result is greater than threshold value then think have car to pass through detection line 2, promptly arrive the optimum detection position, current frame image is imported into wait for next step concatenation in the video processor then.Get access to the grey scale change that continues to detect detection line 1 position and detection line 3 positions behind the image to be spliced again, till judging that vehicle sails out of surveyed area.
Because the relative simple fixation of background image in the parking lot, in order to improve the efficient of detection, present embodiment has been set up the background template storehouse.Utilize the background image of no car to splice every one hour, and upgrade the template in the template base.
After identification had moving object, two picture frames that video camera absorbed spliced when utilizing image hybrid process based on unique point that vehicle is in the optimum detection position.Image mosaic in the present embodiment all adopts the joining method based on model.Promptly adopt as shown in Figure 6 the image split-joint method based on Feature Points Matching during commissioning device in the first time: (left video camera is taken the photograph on the left of image right side and the image that right video camera is taken the photograph in 60 pixel scopes) carries out feature point extraction in known overlapping region earlier, it is right to adopt the gray scale template matching method to find out corresponding match point again, utilizes least square method to obtain the transformation matrix of two width of cloth images.In transformation matrix, find out translational movement between the image and rotation amount and preserve as the model parameter of each binocular video collector, real-time detected image splicing afterwards just according to these two parameters with image translation, rotation, carry out image co-registration again and get final product.
After obtaining complete passage vehicle image, utilize the car plate recognizer to extract the number-plate number.
The output port of video processor links to each other with host computer by network interface, sends information such as time of extracting, car plate to host computer by network interface, and receives the control command that host computer sends, and binocular video collector, video memory are controlled.
Present embodiment also provides construction and installation topological diagram planning function, calculates effective surveyed area of every detecting device according to the radiation parameters of video camera in the binocular video detecting device, and the building structure according to the parking lot provides the construction solution again.
Present embodiment in use cost is low, the real-time height, and the accuracy height, installation and maintenance are simple, possess the good commercial practicality.
Claims (1)
1, video channel wagon detector is characterized in that: comprise binocular video collector, video memory, video processor and network interface;
Described binocular video collector, comprise two video cameras and A the D change-over circuit, the distance between two video cameras is 15 centimetres～30 centimetres; Two video cameras are taken the image in its respective channel respectively, and have certain coincidence visual field, and the width range that overlaps the visual field is 15 centimetres～30 centimetres; Two simulating signals that video camera absorbed change into data image signal by A D change-over circuit;
A the digital signal output interface of D change-over circuit link to each other with the video data input interface of video memory, the passage video digital images information that collects is exported to video memory;
The video data output interface of video memory links to each other with video processor, sends the passage video digital images signal that receives to video processor, and video processor is done analyzing and processing to the binocular video image that receives and extracted relevant information;
The output port of video processor links to each other with host computer by network interface, the various information that video processor is extracted sends to host computer by network interface, and receive the control command that host computer sends, binocular video collector, video memory are controlled.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008202338582U CN201345166Y (en) | 2008-12-26 | 2008-12-26 | Video-based passage vehicle detector |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNU2008202338582U CN201345166Y (en) | 2008-12-26 | 2008-12-26 | Video-based passage vehicle detector |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201345166Y true CN201345166Y (en) | 2009-11-11 |
Family
ID=41276750
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNU2008202338582U Expired - Fee Related CN201345166Y (en) | 2008-12-26 | 2008-12-26 | Video-based passage vehicle detector |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201345166Y (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096591A (en) * | 2014-05-14 | 2015-11-25 | 杭州海康威视数字技术股份有限公司 | Intelligent road monitoring system and method |
| CN105225482A (en) * | 2015-09-02 | 2016-01-06 | 上海大学 | Based on vehicle detecting system and the method for binocular stereo vision |
| EP3121761A1 (en) * | 2015-07-20 | 2017-01-25 | Dura Operating, LLC | A system and method for verifying road position information for a motor vehicle |
| CN106875690A (en) * | 2017-03-01 | 2017-06-20 | 深圳市创为安防有限公司 | For the camera head and its monitoring method of vehicle monitoring |
| CN105096591B (en) * | 2014-05-14 | 2018-06-01 | 杭州海康威视数字技术股份有限公司 | Intelligent road monitoring system and method |
| CN109377768A (en) * | 2018-10-24 | 2019-02-22 | 成都臻识科技发展有限公司 | A kind of parking detection method and system for curb parking |
| CN112116686A (en) * | 2020-09-21 | 2020-12-22 | 猫岐智能科技（上海）有限公司 | Data fusion drawing system and method based on multi-direction light curtain scanning |
| CN113257033A (en) * | 2021-07-01 | 2021-08-13 | 成都宜泊信息科技有限公司 | Parking lot management method and system, storage medium and electronic equipment |
-
2008
- 2008-12-26 CN CNU2008202338582U patent/CN201345166Y/en not_active Expired - Fee Related
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105096591A (en) * | 2014-05-14 | 2015-11-25 | 杭州海康威视数字技术股份有限公司 | Intelligent road monitoring system and method |
| CN105096591B (en) * | 2014-05-14 | 2018-06-01 | 杭州海康威视数字技术股份有限公司 | Intelligent road monitoring system and method |
| EP3121761A1 (en) * | 2015-07-20 | 2017-01-25 | Dura Operating, LLC | A system and method for verifying road position information for a motor vehicle |
| CN106568448A (en) * | 2015-07-20 | 2017-04-19 | 德韧营运有限责任公司 | System and method for verifying road position information for motor vehicle |
| CN105225482A (en) * | 2015-09-02 | 2016-01-06 | 上海大学 | Based on vehicle detecting system and the method for binocular stereo vision |
| CN105225482B (en) * | 2015-09-02 | 2017-08-11 | 上海大学 | Vehicle detecting system and method based on binocular stereo vision |
| CN106875690A (en) * | 2017-03-01 | 2017-06-20 | 深圳市创为安防有限公司 | For the camera head and its monitoring method of vehicle monitoring |
| CN109377768A (en) * | 2018-10-24 | 2019-02-22 | 成都臻识科技发展有限公司 | A kind of parking detection method and system for curb parking |
| CN109377768B (en) * | 2018-10-24 | 2021-10-12 | 成都臻识科技发展有限公司 | Parking detection method and system for roadside parking |
| CN112116686A (en) * | 2020-09-21 | 2020-12-22 | 猫岐智能科技（上海）有限公司 | Data fusion drawing system and method based on multi-direction light curtain scanning |
| CN113257033A (en) * | 2021-07-01 | 2021-08-13 | 成都宜泊信息科技有限公司 | Parking lot management method and system, storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN201345166Y (en) | Video-based passage vehicle detector | |
| CN101510356B (en) | Video detection system and data processing device thereof, video detection method | |
| CN108010375A (en) | The intelligent cruise robot reverse vehicle searching system and method being wirelessly transferred based on high in the clouds | |
| CN102622783B (en) | Multilane free-flow electronic toll collection method based on accurate position match | |
| CN106680290B (en) | Multifunctional detection vehicle in narrow space | |
| CN107380163A (en) | Automobile intelligent alarm forecasting system and its method based on magnetic navigation | |
| CN103413313B (en) | The binocular vision navigation system of electrically-based robot and method | |
| CN111753797B (en) | Vehicle speed measuring method based on video analysis | |
| CN101727748B (en) | Method, system and equipment for monitoring vehicles based on vehicle taillight detection | |
| CN110542898A (en) | Radar group-based vehicle behavior continuous tracking detection system and method | |
| CN111639505B (en) | Hybrid positioning navigation system and method for indoor inspection robot | |
| CN107993456B (en) | Single-way intelligent traffic light control system and method based on end-of-traffic of sidewalk | |
| CN105654732A (en) | Road monitoring system and method based on depth image | |
| CN102915638A (en) | Surveillance video-based intelligent parking lot management system | |
| CN102426785B (en) | Traffic flow information perception method based on contour and local characteristic point and system thereof | |
| CN104217439A (en) | Indoor visual positioning system and method | |
| CN114113118B (en) | Rapid detection device and detection method for subway tunnel lining crack leakage water disease | |
| CN103837087B (en) | Pantograph automatic testing method based on active shape model | |
| CN102789686B (en) | Road traffic flow detecting method based on road surface brightness composite mode recognition | |
| CN106023645A (en) | Photoelectric-and-radar-combination parking space detecting system | |
| CN107065871A (en) | It is a kind of that dining car identification alignment system and method are walked based on machine vision certainly | |
| CN106355895A (en) | Traffic flow and vehicle speed collecting device and method based on infrared matrixes | |
| CN204305214U (en) | Loading goods train state high-definition image system | |
| CN111506069B (en) | All-weather all-ground crane obstacle identification system and method | |
| CN103413449A (en) | Expressway ramp intelligent way directing system and way directing method thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20091111 Termination date: 20131226 |