CN101833117A - Intelligent vehicle safety vision detection device and motion target tracking method - Google Patents
Intelligent vehicle safety vision detection device and motion target tracking method Download PDFInfo
- Publication number
- CN101833117A CN101833117A CN201010179729A CN201010179729A CN101833117A CN 101833117 A CN101833117 A CN 101833117A CN 201010179729 A CN201010179729 A CN 201010179729A CN 201010179729 A CN201010179729 A CN 201010179729A CN 101833117 A CN101833117 A CN 101833117A
- Authority
- CN
- China
- Prior art keywords
- image
- car
- main control
- control computer
- detection device
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Traffic Control Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an intelligent vehicle safety vision detection device and a motion target tracking method. The system mainly comprises a vehicle detection device, a vehicle bottom image scanning device, a license number image acquiring and recognizing module, an image compression and transmission module, an image database module, a main control computer, a box body and the like. The top of the box body is provided with toughened glass, the vehicle bottom image scanning device is arranged below the toughened glass, a camera is connected with the main control computer through a video data output line, wherein the main control computer is internally provided with an image acquiring card; and the intelligent vehicle safety vision detection device is also provided with a license number image acquiring and transmitting part. Meanwhile, a multi-motion target detection and tracking method based on self-adaption nonparametric nuclear density estimation is designed in the invention. The invention has high automatism, high detection speed and wide adaptability, and can realize non-parking detection.
Description
Technical field
The present invention is mainly concerned with the intelligent vehicle safety detection range, refers in particular to a kind of vehicle security vision intelligent detection device in important places such as customs, military base, convention, Air Branch and tracking of moving target of being applicable to.
Technical background
The world today, automobile is massesization more and more, are subject to people's attention day by day at the security monitoring measure of vehicle.Particularly car bombing constantly takes place in various places, and terroristic organization's activity is becoming increasingly rampant.For life, the property safety that ensures the social stability and the people, each state all strengthens the security monitoring measure to important places such as customs, airport, government department, military bases one after another.The terrorist can conceal the sight line of hiding the supervisory personnel in the automobile bottom to explosive mostly.In addition, many personnel of stealing into another country conceal in the bottom of large car to realize illegal entry.Therefore, the bottom of automobile has become the key object of safety inspection.Yet because the restriction of development in science and technology level, inspection method mainly adopts at the bottom of the present car is robot etc. at the bottom of hand-held safety check mirror, the novel car.
The defective of these methods is mainly reflected in:
1, the safety check microscopy is surveyed at the bottom of the use car.The significant disadvantage of this method is that checking efficiency is extremely low, when vehicle is checked, must be with vehicle parking in appointed positions, and hand-held safety check mirror is stretched into the bottom of tested vehicle by special supervisory personnel, judge that by observing at the bottom of the car that the safety check mirror reversal penetrates mirror image whether have at the bottom of the car can attachment.Usually check that a dilly needs 2~3 minutes, check medium-sized or oversize vehicle, required time will be longer.Because there are defectives such as examination scope is imperfect at the bottom of the car, human resources consumption is big in the restriction of safety check mirror itself, this inspection method.
2, adopt novel car at the bottom of robot detect.Camera and radio communication device have been installed in this robot, in use by the site operation personnel by wireless remote controller to car at the bottom of robot send instruction, control the shooting angle of its direction of motion, speed and camera, robot is sent to the image that photographs on the controller display screen in operating personnel's hand.Though at the bottom of this car safety check robot than the hand-held car at the bottom of the safety check mirror had significantly and to have improved, the deficiency of this detection method is: (i) need tested vehicle parking to be checked; The image information that (ii) obtains might be imperfect, the place that dolly does not traverse can occur; (iii) check consuming time longer, compact car be checked for up to 6~8 minutes.
At present, the research and development of domestic R﹠D institution have similar safe examination system based on machine vision, and its price is relatively low, but the function of its realization only is the video information of vehicle bottom when taking automobile by the safety check station with camera, and be presented on the on-the-spot monitor, there is not the memory function of file.And, because what show is dynamic video at the bottom of the car, but not the image at the bottom of the car, so probably miss some very important information because the speed of a motor vehicle is too fast or the staff is absent minded when implementing monitoring, this safety inspection effect to important places such as customs is very unfavorable.And scope only is 1.3 meters at the bottom of the car that this system can photograph, and detection speed only can reach 0.3 meter per second.Defectives such as the product of foreign market is perfect to some extent in function, but still has the cost height, and extensibility is not strong.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, designed a kind of intelligent vehicle safety vision detection device, and proposed the motion target tracking method that adopts in the system.This plant automation degree height, detection speed is fast, precision is high, applied widely, and the vehicle security vision intelligent detection device that can realize not stopping and detect.
Concrete technical solution of the present invention is as follows:
A kind of intelligent vehicle safety vision detection device is characterized in that comprising motor vehicle detector, car base map as scanister, license plate image collection and identification module, compression of images and transport module and view data library module, and main control computer, casing etc.
Described motor vehicle detector is provided with the triggering collection unit, and described triggering collection unit comprises ground induction coil and chromacoder, and described chromacoder is connected on the serial line interface of main control computer.
Described car base map uses the high speed linear array CCD smear camera as scanister, and its video data output line links to each other with the main control computer of built-in high speed image acquisition board, is equipped with higher source luminance simultaneously.
What described license plate image collection and identification module adopted is the high resolving power Array CCD Camera, and the license plate image that obtains by the identification module quick identification.
Described compression of images and transport module, at the bottom of adopting JPEG still image compression standard to the car that collects and license plate image compress, by the TCP/IP network communication interface, the method that utilization CLIENT and Winsock programme and combine has realized the real-time Transmission of image file.
Described view data library module, be responsible at the bottom of the on-the-spot car of sending of each monitoring that receives and the car plate picture manage, and provide basic image retrieval function.Its database kernel comprises that multiple spot source image management, image show control, historical data processing, image retrieval processing etc.
Described motor vehicle detector and car base map all are installed in the described casing as scanister, and the casing top is provided with lamina of septum pellucidum, are beneficial to obtaining of car base map picture.
Described light source cell adopts high frequency high brightness halogen light source.
The multiple mobile object that is based on the estimation of self-adaptation norm of nonparametric kernel density that described method for tracking target adopts detects and tracking.
Compared with prior art, advantage of the present invention just is:
1, vehicle security vision intelligent detection device of the present invention is to detect pick-up unit automatically, need not manually-operated, its automaticity height, detection speed is fast, accuracy of detection is high, applied widely, and can realize the detection of not stopping, therefore improved the reliability of safety check efficient and safety check greatly;
2, vehicle security vision intelligent detection device of the present invention is embedded in the face of land, and is simple in structure, volume is little, does not take too much space, is suitable for the layout under the various environment, and its sensing range is wide, is fit to the various vehicles of different size.
3, vehicle security vision intelligent detection device of the present invention not only can detect vehicle bottom image and car plate picture in real time, can also store by main control computer, operation such as network inquiry, enrich the content of real-time safety check, further improved the reliability of safety check.
4, the present invention proposes to set up the background model of estimating based on the self-adaptation norm of nonparametric kernel density, and the bandwidth of Density Estimator device is definite by the Probability Characteristics of sampled point, estimation point, can effectively suppress to disturb background model accuracy height.
5, the present invention has designed object matching situation between " coupling matrix " analysis frame, and according to matching result, proposes target state is divided into four classes, takes corresponding tracking strategy respectively.Algorithm has been realized the tracking of multiple mobile object under the different motion situation, and can obtain target trajectory.
Description of drawings
Fig. 1 is a general structure synoptic diagram of the present invention;
Fig. 2 is an imaging device installation sectional view of the present invention;
Fig. 3 is that imaging device of the present invention is installed vertical view;
Fig. 4 is a vehicle detector connection diagram of the present invention;
At the bottom of Fig. 5 is car of the present invention, license plate image imaging processing process flow diagram;
Fig. 6 is car plate identification process figure of the present invention;
Fig. 7 is the retrieval flow figure based on picture material of the present invention;
Fig. 8 is multi-target detection of the present invention, tracking basic flow sheet;
Fig. 9 is a target detection process flow diagram of the present invention.
The number in the figure explanation:
1, tempered glass 2, casing 3, main control computer
4, Array CCD Camera optical cable 5, Array CCD Camera 6, bracing frame
7, support post 8, linear array CCD camera optical cable 9, casing external tapping
10, linear array CCD camera 11, halogen light source 12, ground
13, photoelectric sensor
Embodiment
Embodiment 1:
Below with reference to the drawings and specific embodiments the present invention is described in further details.
As shown in Figure 1.Intelligent vehicle safety vision detection device of the present invention, mainly comprise main control computer 3, be embedded in the casing 2 on the face of land, the car base map that is installed on casing 2 inside is provided with tempered glass 1 (and tempered glass 1 flushes installation with the face of land), is built in ground license plate image harvester 4 (this device is riveted on the face of land by column 7, with casing 2 at a distance of about 2.5 meters) as the top of harvester 10, light source 11, casing 2.The car base map links to each other with main control computer 3 by its video data output line as harvester 10.Casing 2 adopts iron or steel design, and the tempered glass 1 at its top has guaranteed the bearing capacity and the crushing resistance of equipment integral.License plate image harvester 5 calculates 3 by its video output cable with the master control of built-in scan video image pick-up card and links to each other, and license plate image harvester 5 adopts high-resolution face smear camera.
Fig. 2,3 is imaging device installation cut-open view and vertical view.The imaging video camera is installed in the casing 2 at the bottom of the car, and casing 2 highly is designed to 450 millimeters, and the distance of the tempered glass x at camera lens top and top is 300 millimeters.
In order to obtain the car base map picture of high speed, high resolution, the present invention has used the high speed linear array CCD video camera, adds camera lens.Light source cell x adopts the high frequency halogen light source, and its frequency can reach KHz.In the present embodiment, casing 2 outsides are equiped with the triggering collection unit, and this triggering collection unit comprises ground induction coil and chromacoder, and ground induction coil is installed in casing 2 outsides, and chromacoder is arranged on the serial line interface of main control computer 3.
Fig. 4 is the vehicle detector connection diagram.The present invention is based on one-chip computer module is kernel control module, has finished to the collection of ground induction coil trigger pip and by serial line interface and has transmitted data to host computer.Design an I/O mouth that takies single-chip microcomputer and gathered the signal of ground induction coil.Acquisition mode is the mode of inquiring about in real time, just sends corresponding data to host computer in case detect the variation of I/O mouth level.On communication modes, system adopts the RS485 communication mode, and the signal demand that single-chip microcomputer sends from serial ports converts the RS485 level of standard to through a serial ports level transferring chip.
At the bottom of Fig. 5 is car, license plate image imaging processing process flow diagram.After system powered on, if adopt automatic drainage pattern, ground induction coil was triggered, and at the bottom of then starting image pick-up card and beginning car, the image acquisition of car plate, after vehicle was fully by scanister at the bottom of the car, the car base map looked like to be saved, shows and handles.At this moment, if note abnormalities, then start the abnormal conditions warning device, whether if no abnormal, it is ready then to detect the network connection, ready then to server transmission monitoring record, withdraws from supervisory system at last.
Fig. 6 is car plate identification process figure.To the license plate image that obtains, what at first carry out is the image pre-service, comprise that figure image intensifying and edge of image detect, wherein can adopt gray scale linearization, histogram equalization and binaryzation to carry out the figure image intensifying, use Roberts operator, Sobel operator, Prewitt operator and Canny operator and carry out the edge of image detection.Then carry out the location of car plate, can take methods such as linear edge detection, threshold value iteration, neural network, gray scale, color characteristic to position.Behind the location, just can carry out cutting apart of character, at first adopt sciagraphy, gravity model appoach, Hough method of changing to tilt to correct, the back uses the Return Law, template matching method to cut apart.Final step promptly is the identification of character, can adopt pattern-recognition, Bayesian statistics pattern-recognition, the structure of structure to combine with statistics, method such as artificial neural network and vignette template matches.
Fig. 7 is the retrieval flow figure based on picture material of the present invention.The car plate that system obtains, car base map picture are all deposited in the image data base by compression, transmission, at this moment, the user inquires about by man-machine interface, the feature of input and the feature of extracting from image data base are carried out similarity comparison, coupling, if the result is similar, then provide feedback information, and provide result for retrieval at display interface.
Fig. 8 is multi-target detection of the present invention, tracking basic flow sheet.Current frame image to input carries out the detection of multiple mobile object, at first determines the target area to be followed the tracks of by multiple mobile object wherein.
Fig. 9 be to the input current frame image, with each pixel in the image as estimation point, application self-adapting norm of nonparametric kernel density method of estimation is calculated the probability of current frame pixel point, as probability less than threshold value T, then this pixel is an impact point, and the set that obtains all impact points is detected target.
Claims (3)
1. intelligent vehicle safety vision detection device and motion target tracking method, it is characterized in that, comprise motor vehicle detector, car base map as scanister, license plate image collection and identification module, compression of images and transport module and view data library module, and main control computer, casing etc.Described motor vehicle detector is provided with the triggering collection unit, and described triggering collection unit comprises ground induction coil and chromacoder, and described chromacoder is connected on the serial line interface of main control computer.
2. according to claim 1, described car base map uses the high speed linear array CCD smear camera as scanister, and its video data output line links to each other with the main control computer of built-in high speed image acquisition board, is equipped with higher source luminance simultaneously.What described license plate image collection and identification module adopted is the high resolving power Array CCD Camera, and the license plate image that obtains by the identification module quick identification.Described compression of images and transport module, at the bottom of adopting JPEG still image compression standard to the car that collects and license plate image compress, by the TCP/IP network communication interface, the method that utilization CLIENT and Winsock programme and combine has realized the real-time Transmission of image file.Described view data library module, be responsible at the bottom of the on-the-spot car of sending of each monitoring that receives and the car plate picture manage, and provide basic image retrieval function.Its database kernel comprises that multiple spot source image management, image show control, historical data processing, image retrieval processing etc.Described motor vehicle detector and car base map all are installed in the described casing as scanister, and the casing top is provided with lamina of septum pellucidum, are beneficial to obtaining of car base map picture.Described light source cell adopts high frequency high brightness halogen light source.
3. according to claim 1, the multiple mobile object that is based on the estimation of self-adaptation norm of nonparametric kernel density that described method for tracking target adopts detects and tracking.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010179729A CN101833117A (en) | 2010-05-24 | 2010-05-24 | Intelligent vehicle safety vision detection device and motion target tracking method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201010179729A CN101833117A (en) | 2010-05-24 | 2010-05-24 | Intelligent vehicle safety vision detection device and motion target tracking method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN101833117A true CN101833117A (en) | 2010-09-15 |
Family
ID=42717255
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201010179729A Pending CN101833117A (en) | 2010-05-24 | 2010-05-24 | Intelligent vehicle safety vision detection device and motion target tracking method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101833117A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2449374C1 (en) * | 2011-04-13 | 2012-04-27 | Федеральное государственное унитарное предприятие "Таганрогский научно-исследовательский институт связи" (ФГУП "ТНИИС") | System for monitoring operation of vehicles |
CN103207997A (en) * | 2013-04-15 | 2013-07-17 | 浙江捷尚视觉科技有限公司 | Kernel density estimation-based license plate character segmentation method |
CN104112299A (en) * | 2014-07-02 | 2014-10-22 | 陕西科技大学 | Intelligent parking charging system and charging method thereof |
CN107122732A (en) * | 2017-04-25 | 2017-09-01 | 福州大学 | The quick license plate locating method of high robust under a kind of monitoring scene |
CN107677683A (en) * | 2017-10-26 | 2018-02-09 | 北京华力兴科技发展有限责任公司 | Train under-pan examining device |
CN107808392A (en) * | 2017-10-31 | 2018-03-16 | 中科信达(福建)科技发展有限公司 | The automatic method for tracking and positioning of safety check vehicle and system of open scene |
CN108538060A (en) * | 2018-04-13 | 2018-09-14 | 上海工程技术大学 | A kind of intelligence based on vehicle-mounted back vision camera is broken rules and regulations monitoring method and system |
CN112308073A (en) * | 2020-11-06 | 2021-02-02 | 中冶赛迪重庆信息技术有限公司 | Method, system, equipment and medium for identifying loading and unloading transshipment state of scrap steel train |
CN113418891A (en) * | 2021-07-14 | 2021-09-21 | 青岛大学 | Terahertz ground detection system for detecting safety of bottom of vehicle and detection method thereof |
-
2010
- 2010-05-24 CN CN201010179729A patent/CN101833117A/en active Pending
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
RU2449374C1 (en) * | 2011-04-13 | 2012-04-27 | Федеральное государственное унитарное предприятие "Таганрогский научно-исследовательский институт связи" (ФГУП "ТНИИС") | System for monitoring operation of vehicles |
CN103207997A (en) * | 2013-04-15 | 2013-07-17 | 浙江捷尚视觉科技有限公司 | Kernel density estimation-based license plate character segmentation method |
CN103207997B (en) * | 2013-04-15 | 2017-02-08 | 浙江捷尚视觉科技股份有限公司 | Kernel density estimation-based license plate character segmentation method |
CN104112299A (en) * | 2014-07-02 | 2014-10-22 | 陕西科技大学 | Intelligent parking charging system and charging method thereof |
CN107122732B (en) * | 2017-04-25 | 2019-12-31 | 福州大学 | High-robustness rapid license plate positioning method in monitoring scene |
CN107122732A (en) * | 2017-04-25 | 2017-09-01 | 福州大学 | The quick license plate locating method of high robust under a kind of monitoring scene |
CN107677683A (en) * | 2017-10-26 | 2018-02-09 | 北京华力兴科技发展有限责任公司 | Train under-pan examining device |
CN107808392A (en) * | 2017-10-31 | 2018-03-16 | 中科信达(福建)科技发展有限公司 | The automatic method for tracking and positioning of safety check vehicle and system of open scene |
CN107808392B (en) * | 2017-10-31 | 2023-04-07 | 中科信达(福建)科技发展有限公司 | Automatic tracking and positioning method and system for security check vehicle in open scene |
CN108538060A (en) * | 2018-04-13 | 2018-09-14 | 上海工程技术大学 | A kind of intelligence based on vehicle-mounted back vision camera is broken rules and regulations monitoring method and system |
CN112308073A (en) * | 2020-11-06 | 2021-02-02 | 中冶赛迪重庆信息技术有限公司 | Method, system, equipment and medium for identifying loading and unloading transshipment state of scrap steel train |
CN112308073B (en) * | 2020-11-06 | 2023-08-25 | 中冶赛迪信息技术(重庆)有限公司 | Method, system, equipment and medium for identifying loading and unloading and transferring states of scrap steel train |
CN113418891A (en) * | 2021-07-14 | 2021-09-21 | 青岛大学 | Terahertz ground detection system for detecting safety of bottom of vehicle and detection method thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101833117A (en) | Intelligent vehicle safety vision detection device and motion target tracking method | |
CN102915638A (en) | Surveillance video-based intelligent parking lot management system | |
CN109291841B (en) | Intelligent investigation patrol police car | |
CN100419813C (en) | Omnibearing visual sensor based road monitoring apparatus | |
CN204117361U (en) | Based on the Intelligent Parking Peccancy Monitoring System of 360 degree of panoramic cameras | |
CN101762327B (en) | Infrared temperature monitoring method and system of electrified railway contact network | |
CN103646250B (en) | Pedestrian monitoring method and device based on distance image head and shoulder features | |
CN107845289A (en) | A kind of Roadside Parking management and charging system based on height digital video array | |
CN210899299U (en) | Tunnel monitoring system | |
CN108710827B (en) | A kind of micro- police service inspection in community and information automatic analysis system and method | |
CN113985400B (en) | Monitoring and alarming system and method for foreign matters outside airport runway | |
CN107134172A (en) | Parking monitors identifying system and its method | |
CN103646550A (en) | Intelligent vehicle license plate recognition system | |
CN101699531A (en) | Infrared correlation identification alarm system and identification method thereof | |
CN107681784B (en) | Steel tower ground wire monitoring system and method based on multisource data fusion | |
CN106355895A (en) | Traffic flow and vehicle speed collecting device and method based on infrared matrixes | |
CN103321129A (en) | Remote pavement construction diagnosis system and method based on infrared thermal image of third-generation (3G) network | |
CN104537786A (en) | Infrared remote image identification and alarm device for transformer station | |
CN207676502U (en) | A kind of Roadside Parking managing device of included charging pile | |
KR101368757B1 (en) | Parked vehicle monitoring system and method | |
CN207909351U (en) | A kind of guiding of parking stall and reverse vehicle searching system | |
CN110866462A (en) | Behavior recognition system and method integrated in intelligent police car | |
CN201654933U (en) | Infrared associative recognition alarm system | |
CN207704502U (en) | Multifunctional running records system | |
CN203376876U (en) | Parking-lot free parking stall data-collection system based on image analysis |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20100915 |