CN1920896A - Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate - Google Patents

Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate Download PDF

Info

Publication number
CN1920896A
CN1920896A CN 200610068430 CN200610068430A CN1920896A CN 1920896 A CN1920896 A CN 1920896A CN 200610068430 CN200610068430 CN 200610068430 CN 200610068430 A CN200610068430 A CN 200610068430A CN 1920896 A CN1920896 A CN 1920896A
Authority
CN
China
Prior art keywords
vehicle
car plate
image
light filling
rgb
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
Application number
CN 200610068430
Other languages
Chinese (zh)
Other versions
CN100449576C (en
Inventor
朱传琴
王亚盛
许萍
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Conway Communication Technology Co Ltd
Original Assignee
Individual
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CNB2006100684302A priority Critical patent/CN100449576C/en
Publication of CN1920896A publication Critical patent/CN1920896A/en
Application granted granted Critical
Publication of CN100449576C publication Critical patent/CN100449576C/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Traffic Control Systems (AREA)
  • Image Processing (AREA)

Abstract

The invention relates to a license plate real-time image detecting and false-truth recognize system, which can automatically adjust and fill color, wherein the invention comprises front camera device and information processing center, which use public/special communication network to transmit data; the front camera device comprises RGB color filling lighting lamp, color digit camera used to detect, RGB splitter, lighting lamp, camera control circuit, color temperature analysis comparer, image analyzer and algorism logic, front database, A/D converter, image collector, CPU central processor, and digit compression transmitter; the information processing center comprises web special server, information database, information processing platform, compression-decompression and transmission interface. The invention can real-time detect the license plate, type, and their match result, and recognize the false and truth.

Description

The car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling
Technical field
The present invention relates to the image recognition and the monitoring field of electronic information technology, especially relate to a kind of car plate vehicle real-time image monitoring and genuine-fake identification system of automatic color regulating light filling.
Background technology
False car plate, the many cars of one car plate use, the situation of stealing or borrowing car plate to use happens occasionally, traffic safety and social security in these illegal activities serious harms, it is the emphasis of vehicle supervision department's control and strike, and the video monitoring system of traffic police department use at present, only can discern car plate, the vehicle that violates the traffic regulations is taken pictures, and true and false for car plate, car plate and vehicle are not inconsistent, or one violation of law such as many cars of car plate use can't differentiate, also exist simultaneously under the bad situation of light, shooting image blurring unclear such as can't discern at technical matters, the discrimination of car plate is lower, therefore, traffic police department presses for a kind of vehicular traffic video monitoring system of multiple functional and round-the-clock use, to satisfy the demand of traffic safety and social security.
Technology contents
Purpose of the present invention is exactly in order to solve present the deficiencies in the prior art, a kind of traffic administration needs that can satisfy the traffic police to vehicle are provided, ensure normal traffic order and safety, the various robber's cars of severe strike, walk private car, the illegal activities that a fake license and a car plate are used more realize the car plate to various vehicles, and the situations such as legitimacy of vehicle and car plate vehicle coupling are carried out the car plate vehicle real-time image monitoring and the genuine-fake identification system of the automatic color regulating light filling of real-time video monitoring and identification.
For achieving the above object, the present invention has taked following technical scheme:
The present invention is divided into front end picture pick-up device and information processing centre two parts, adopt the public/private communication network to carry out data transmission between front end picture pick-up device and the information processing centre, the front end picture pick-up device comprises RGB toning light filling illuminating lamp, monitor and use the colored digital video camera, the RGB spectroscope, illuminating lamp and shooting control circuit, colour temperature is analyzed comparing unit, graphical analysis and algorithm logic, front end data storehouse, A/D conversion and image acquisition, cpu central processing unit and digital compression hop, information processing centre comprises the Web private server, document data base, information processing platform system and Compress softwares and communication interface.
In the front end picture pick-up device, RGB toning light filling illuminating lamp is by redness, green, blue and infrared four LED power models constitute, control the working current of four kinds of LED power models respectively by illuminating lamp and shooting control circuit, produce the compensation light of different-colour and illumination, RGB spectroscope and colour temperature are analyzed comparing unit and mainly the colour temperature and the illumination of background light are carried out Measurement and analysis, and compare with standard color temperature and illumination and to calculate the parameter request that need carry out light filling, illuminating lamp is also controlled the location parameter x of two supervision with the colored digital video camera by 6 DC servo-motor with the shooting control circuit, y, α, consistance with the location parameter of the location parameter that guarantees shooting and vehicle template, improve discrimination, graphical analysis and algorithm logic and A/D conversion mainly are that video image is gathered and analog/digital signal conversion with image acquisition, obtain the number-plate number by pre-service and logical calculated, elementary parameters such as vehicle trade mark and vehicle specification send information processing centre to and carry out detailed comparative analysis and processing, true and false car plate and vehicle are screened, cpu central processing unit is responsible for handling information transmission between the various piece and processing in the front end picture pick-up device, and communicate by public/private communication network and information processing centre, the main stored programme file in front end data storehouse, basis vehicle template data and the video image that detects kept in, because the byte number that vision signal takies is very big, be sent to information processing centre after need compressing by the digital compression transmission circuit, to guarantee high-speed transfer, improve the speed of pickup image and information transmission.
The inner Web private server of setting up of information processing centre mainly finish with outside the writing to each other of template data between the regional traffic police department and vehicle dynamic data, to form provincial or national network system, document data base is mainly stored the vehicle Template Information in the local zone, video monitoring information and system control program etc., Compress softwares and communication interface are finished the compression and the decompress(ion) of the transmission data between information processing centre and the front end picture pick-up device and are set up communication link, information processing platform system is the package of information processing centre, be responsible for the information communication between the agreement each several part, analysis result is proposed to handle suggestion, propose diagram when finding false car plate or false car are arranged and report to the police and audible alarm.
In the front end picture pick-up device, be provided with two supervision colored digital video camera A and B, be installed in respectively on the bracing frame of top, cross-road both sides certain altitude, constitute certain x with the vehicle that is monitored, the location parameter of y distance and α angle can be adjusted within the specific limits, catching vehicle by automatic control tracking is optimum condition with angle and the focal length that guarantees production, the vehicle image of production is divided into car plate, the trade mark and three parts of bassinet structure of before leaning to one side, by cutting apart carrying out framing and necessity respectively after the three part decomposition after the image pre-service, the basic car plate template of contrast front end data library storage, vehicle template and trade mark template, adopt optimum algorithm to carry out edge identification respectively, send information processing centre to after recognition result is compressed and carry out true and false identification.
In illuminating lamp and shooting control circuit, by the RGB spectroscope bias light is decomposed into red R, three kinds of basic monochromatic light of green G and blue B, measure three kinds of monochromatic substantially light intensity analog quantitys by rgb light ionization meter unit, three kinds of monochromatic light intensities that will record by the A/D change-over circuit become digital quantity, analyze comparing unit by colour temperature and analyze comparison with the colour temperature and the illumination of the standard background light of shooting needs, calculating needs compensate for background light illumination and three kinds of monochromatic light intensity components, the standard color temperature of the background light of assurance shooting is about 5250K, if also need to carry out the infrared light light filling to guarantee the pickup quality at night night.
Design has controlled LED constant flow module R, controlled LED constant flow module G, controlled LED constant flow module B and four kinds of independent current source modules of controlled LED constant flow module RL, provide the constant current power supply to four kinds of LED power models respectively, by the instruction of dynamic current and trigger control circuit, adopt digital signal to control the work of constant current source module respectively according to cpu central processing unit.
Vehicle identification comprised car plate is discerned, brand mark identification and three parts of vehicle structure profile identification, car plate is adopted framing, characters on license plate is cut apart, the sane Hausdorff distance calculating method of Canny operator edge detection and self-adaptation carries out Recognition of License Plate Characters, to improve the discrimination of characters on license plate, brand mark is carried out framing, separated graphics is cut apart, cut apart and needn't carry out figure row graph, adopt Canny operator edge detection and the sane Hausdorff distance calculating method of self-adaptation to carry out brand mark identification, two location parameter x that supervision is being set with colored digital video camera A and B about employing, the vehicle structure image of leaning to one side before two is set respectively in the scope of y and α, and write down its location parameter value, by image pre-service and Wavelet Edge Detection, the proper vector similarity is calculated and is carried out location parameter with the vehicle template data of database and proofread and correct back and relatively, determines the brand and the specifications and models of vehicle.
It is by comparing with database vehicle template information that true and false vehicle and car plate are screened, determine whether car plate conforms to the vehicle specification model, if be not inconsistent then be judged to be false car plate or false car, when vehicle conforms to car plate, then the vehicle data of producing according to each surveillance of a specified time in the zone compares, if do not have same vehicle then be judged to be the Zhen Chezhen car plate, if the same vehicle monitor record is arranged, then for twice or repeatedly record carries out the mistiming and the region road differential is analysed, calculate same vehicle in different time and the rationality that different regions occur, judge whether this vehicle is that a plurality of same model vehicles of car plate use.
The vehicle template record adopt same about two supervision with colored digital video camera A and B, front upper side at vehicle carries out the vehicle image setting, in the assigned position parameter x, two vehicle front side body structure images about producing respectively under y and the α numerical value prerequisite, car plate and brand mark image, adopt Canny operator edge detection and the sane Hausdorff distance calculating method of self-adaptation to determine the image digitization amount parameter value of car plate and brand mark, by the image pre-service, Wavelet Edge Detection and proper vector similarity are calculated and are determined vehicle structure image digitization amount parameter, these car plates with vehicle, the digital quantity parameter of brand mark and vehicle structure is stored in as the template parameter of this vehicle in the document data base of information processing centre as the vehicle template parameter, is used for monitoring vehicle is compared and judges.
The invention has the beneficial effects as follows: can effectively discern, satisfy the demand of traffic police, stop vehicle false-trademark etc., satisfy the demand of traffic safety and social security vehicle monitoring to vehicle.
Description of drawings
Fig. 1 is a system construction drawing of the present invention;
Fig. 2 is a vehicle characteristics information acquisition system structural drawing of the present invention;
Fig. 3 is toning light filling and shooting control circuit system structural drawing;
Fig. 4 is a vehicle characteristics sample collection process flow diagram;
Fig. 5 is that vehicle characteristics is gathered the identifying process flow diagram;
Fig. 6 is true and false vehicle discriminant analysis journey process flow diagram;
Fig. 7 is the RGB spectroscopic measurements and compares calculation flow chart.
Wherein, RGB toning light filling illuminating lamp 1 monitors with colored digital video camera 2, RGB spectroscope 3, illuminating lamp and shooting control circuit 4, colour temperature is analyzed comparing unit 5, graphical analysis and algorithm logic 6, front end data storehouse 7, A/D conversion and image acquisition 8, cpu central processing unit 9, digital compression transmission 10, public/private communication network 11, Web private server 12, document data base 13, information processing platform system 14, Compress softwares and communication interface 15 are travelled or static vehicle 16, monitor with colored digital video camera A 17, monitor with colored digital video camera B 18, follow the tracks of and catch vehicle 19, image pre-service 20, car plate location 21, feature pixel location 22, pixel dividing processing 23, edge algorithms identification 24, characters on license plate cuts apart 25, license plate number identification 26, data analysis contrast 27, database vehicle template 28, the result exports 29, rgb light ionization meter unit 30, A/D change-over circuit 31,525nm green light LED power model 32, controlled LED constant current source module G 33,640nm red-light LED power model 34, controlled LED constant current source module R 35,465nm blue-ray LED power model 36, controlled LED constant current source module B 37,940nm infrared LED power model 38, controlled LED constant current source module RL 39, dynamic current and trigger control circuit 40, photoelectric coding change-over circuit 41, motor-driven 42,6 DC servo-motor 43 of controller of making a video recording in advance, current image frame 44, relatively have or not vehicle 45 with former frame, former frame image 46, the target area is clear denys 47, automatically focal length adjusts 48, the bias light colour temperature is suitable 49, light filling illumination control 50, target lock-on 51, brand mark captures 52, lean to one side to capture 53 before the three-dimensional localization, car plate captures 54, and the characteristic area location finishes 55, characteristic area location 56, characteristic area classification 57, brand mark image 58, vehicle front side body structure image 59, license plate image 60, gradient binaryzation 61, consecutive frame difference analysis 62, background rebuilds 63, trade mark and feature cut apart 64, gray processing and smoothing processing 65, Canny operator edge detection 66, Wavelet Edge Detection 67, the sane Hausdorff distance calculation 68 of self-adaptation, the proper vector similarity calculates 69, brand recognition 70, and compute euclidian distances D 71, location parameter x, y, α revises 72, database template parameter 73, vehicle judges 74, vehicle model specification 75, the number-plate number 76, database information of vehicles 77, call database information 78, the car plate model conforms to 79, false car or false car plate 80, database monitoring data 81, have or not similar vehicle 82, true car 83, the mistiming analyzes 84, and the region road differential analyses 85, the speed of a motor vehicle calculates 86, it denys 87 that the speed of a motor vehicle exceeds standard, vehicle background light 88, lens light gathering 89, prism RGB beam split 90, RGB monochromatic light CCD receives 91, is transformed into RGB analog electrical signal 92, A/D conversion output RGB digital signal 93, the RGB digital component 94 of benchmark colour temperature, RGB digital component 95, illuminance digital quantity 96, reference light illumination digital quantity 97, the benchmark light intensity compares 98, benchmark illumination compares 99, calculates RGB difference component 100, calculates illuminance measures of dispersion 101, calculate the controlled LED constant current of RGB ratio 102, start infrared light filling 103, calculate the controlled LED steady current 104 of RGB, calculate the controlled LED steady current 105 of RL.
Embodiment
Below in conjunction with accompanying drawing and embodiment the present invention is made further sets forth in detail.
Embodiment:
Fig. 1 is a system construction drawing of the present invention, the present invention is divided into front end picture pick-up device and information processing centre two parts, adopt public/private communication network 11 to carry out data transmission between front end picture pick-up device and the information processing centre, the front end picture pick-up device comprises RGB toning light filling illuminating lamp 1, monitor with colored digital video camera 2, RGB spectroscope 3, illuminating lamp and shooting control circuit 4, colour temperature is analyzed comparing unit 5, graphical analysis and algorithm logic 6, front end data storehouse 7, A/D conversion and image acquisition 8, cpu central processing unit 9 and digital compression are transmitted 10 parts, and information processing centre comprises Web private server 12, document data base 13, information processing platform system 14 and Compress softwares and communication interface 15.
RGB toning light filling illuminating lamp 1 is by redness, green, blue and infrared four LED power models constitute, the colour temperature and the illumination of the background light of 3 pairs of vehicles of RGB spectroscope are measured, with image acquisition 8 circuit the background light conversion of signals of measuring is become digital quantity by the A/D conversion, analyzing comparing unit 5 by colour temperature compares the standard color temperature stored in the background light digital quantity measured and the front end data storehouse 7 and illumination and calculates the parameter request that need carry out light filling, control the working current of four kinds of LED power models of RGB toning light filling illuminating lamp respectively by illuminating lamp and shooting control circuit 4, produce the compensation light of different-colour and illumination.
Illuminating lamp is controlled two supervision location parameter x with colored digital video camera 2 with shooting control circuit 4 by 6 DC servo-motor, y, and α is with the location parameter that the guarantees shooting consistance with the location parameter of vehicle template, raising discrimination.
Two supervision are changed with image acquisition 8 by A/D with the vehicle image of colored digital video camera 2 productions and are converted digital quantity to, instruction according to cpu central processing unit 9, carry out the image pre-service by graphical analysis and algorithm logic Unit 6, image segmentation and identification, with the number-plate number of confirming, elementary parameters such as vehicle trade mark and vehicle specification are carried out digital compression transmission 10 backs and are carried out detailed comparative analysis and processing by the information vehicle platform system 14 that public/private communication network 11 sends to information processing centre, and true and false car plate and vehicle are screened.
Web private server 12 mainly finish and the traffic police department in outer area between vehicle template data and the writing to each other of vehicle dynamic data, to form provincial or national network system, vehicle Template Information in the document data base 13 main storage local zones, video monitoring information and system control program etc., Compress softwares and communication interface 15 are finished the compression and the decompress(ion) of the transmission data between information processing centre and the front end picture pick-up device and are set up communication link.
Fig. 2 is a vehicle characteristics information acquisition system structural drawing of the present invention, enter supervision with colored digital video camera A 17 with after monitoring the monitoring range of using colored digital video camera B 18 in travels down or static vehicle 16, monitor to follow the tracks of with colored digital video camera B 18 by illuminating lamp and the control of shooting control circuit 4 and catch vehicle 19 with colored digital video camera A 17 and supervision, the image information of by graphical analysis and algorithm logic Unit 6 vehicle 19 is caught in tracking according to cpu central processing unit 9 instruction is carried out image pre-service 20 and is carried out car plate location 21 respectively and comprise brand mark and the feature pixel of the vehicle structure image of before leaning to one side locatees 22, by graphical analysis and algorithm logic Unit 6 license plate image 60 of locating is carried out computings such as Character segmentation 25 and Canny operator edge detection 66, carry out accurate license plate number identification 26, the feature pixel image of locating is carried out computings such as necessary pixel dividing processing 23 and Wavelet Edge algorithm identified 24 by graphical analysis and algorithm logic Unit 6, carry out accurate brand mark and vehicle structure image recognition.
Transmitting Unit 10 through the vehicle parameter information after license plate number identification 26 and edge algorithms identification 24 identifications by digital compression carries out being transferred to information processing centre after digitalization compressed is handled, call the corresponding database vehicle template 28 of document data base 13 stored by information processing platform system 14 and carry out comprehensive data analysis contrast 27, draw the true and false result of determination output 29 of this vehicle.
Fig. 3 is toning light filling and shooting control circuit system structural drawing, after gathering the background light that is monitored vehicle, RGB spectroscope 3 is decomposed into ruddiness R, three kinds of homogeneous beams of green glow G and blue light B, convert analog electrical signal to by the high-sensitive CCD element in the rgb light ionization meter unit 30, process A/D change-over circuit 31 is transported to colour temperature analysis comparing unit 5 after converting digital electric signal to, standard background light illumination 1000Lux and colour temperature 5250K that the contrast shooting needs analyze comparison and calculating, the Parameters Transformation that needs light filling that to calculate by the control of cpu central processing unit becomes the control electrical quantity of dynamic steady current and trigger control circuit 40, respectively to controlled LED constant current source module R 35, controlled LED constant current source module G 33, controlled LED constant current source module B 37 and controlled LED constant current source module RL 39 carry out continuous current and set and trigger sequential control, 640nm red-light LED power model 34,525nm green light LED power model 32, the working current of 465nm blue-ray LED power model 36 and 940nm infrared LED power model 38, luminous intensity and luminous sequential are subjected to controlled LED constant current source module R 35 respectively, controlled LED constant current source module G 33, infrared light filling illumination is carried out in the control of controlled LED constant current source module B37 and controlled LED constant current source module RL 39 when too dark for night or vehicle background light.
Cpu central processing unit 9 is according to monitoring with colored digital video camera A 17 and monitoring that following the tracks of the image information situation of catching vehicle 19 with colored digital video camera B 18 carries out graphical analysis by graphical analysis and algorithm logic 6 circuit, determining by cpu central processing unit 9 calculating need be to monitoring with colored digital video camera A 17 and the camera distance x of supervision with colored digital video camera B 18, y, camera position such as angle [alpha] and focal length parameter, location parameter is encoded and the electrical quantity conversion by photoelectric coding change-over circuit 41, carry out position and angular setting by 6 DC servo-motor 43 of motor-driven and 42 controls of shooting controller, to guarantee to monitor with colored digital video camera A 17 and to monitor the camera position parameter x of using colored digital video camera B 18, y, angle [alpha] reaches optimum condition, directly control supervision with colored digital video camera A 17 and the focal length that monitors with colored digital video camera B 18 by coded signal, be stored in the front end data storehouse 7 by adjusted optimal position parameters.
Fig. 4 is a vehicle characteristics sample collection process flow diagram, for monitoring that the consecutive image of recording with colored digital video camera A 17 carries out image frame acquisitions according to the speed of per second collection 10-15 frame, current image frame 44 is gathered the back and is responsible for adopting rapid image difference shadow method and former frame image 46 to compare by graphical analysis and algorithm logic Unit 6, if relatively have or not vehicle 45 to judge no vehicle in the monitoring ranges then continue the next frame image is compared by former frame, if vehicle arranged then determine the target area of vehicle clear not 47, if it is unintelligible then assign instruction by cpu central processing unit 9 and have motor-driven and the control of shooting controller 42 to monitor with colored digital video camera A 17 or monitor to carry out automatic focal length adjustment 48 with colored digital video camera B 18, then carry out the judgement of bias light colour temperature suitable 49 after the target area is clear, when determining that by 5 calculating of colour temperature analysis comparing unit the bias light colour temperature is inappropriate, carry out light filling illumination control 50 by dynamic current and trigger control circuit 40, carry out target lock-on 51 after the bias light colour temperature is suitable, when target lock-on 51, carry out brand mark respectively and capture 52, lean to one side before the three-dimensional localization to capture 53 and car plate capture 54, at this moment the characteristic area of vehicle location finishes 55.
And supervision is carried out vehicle characteristics sample collection flow process with colored digital video camera B 18 and monitor that usefulness colored digital video camera A's 17 is identical.
Fig. 5 is that vehicle characteristics is gathered the identifying process flow diagram, in vehicle characteristics sample collection process, with the characteristic area classification 57 behind the characteristic area location 56 is trade mark standard picture 58, vehicle front side body structure image 59 and license plate image 60 3 classes, license plate image 60 through the location is carried out characters on license plate cut apart 25, then adopt Canny operator edge detection 66 and serial range conversion, obtain changing image apart from the limit, each grey scale pixel value of distance transformed image equals the minimum distance of this pixel to object edge, adopt sane Hausdorff distance calculation 68 methods of the best self-adaptation of present Figure recognition effect to calculate the Hausdorff distance, carry out license plate number identification 26 with this result of calculation, the discrimination that employing Canny operator edge detection and sane Hausdorff distance calculation 68 methods of self-adaptation are carried out car plate and the identification of vehicle brand mark is the highest.
Brand mark image 58 carries out trade mark and feature cuts apart 64, its partition principle is that row graph does not carry out figure and cuts apart, interrupted image carries out figure according to interrupted piece to be cut apart, figure is cut apart the back and is adopted Canny operator edge detection 66 and serial range conversion, obtain changing image apart from the limit, each grey scale pixel value of distance transformed image equals the minimum distance of this pixel to object edge, adopt sane Hausdorff distance calculation 68 methods of the best self-adaptation of present Figure recognition effect to calculate the Hausdorff distance, carry out brand mark identification 70 with this result of calculation.
Two location parameter x that supervision is being set with colored digital video camera B 18 with colored digital video camera A 17 and supervision about employing, two vehicle front side body structure images 59 about having produced respectively in the scope of y and α, and write down its location parameter value, when being arranged, adopt lighter shade gradient binaryzation 61 methods to eliminate the influence of part shade, and adopt fast background 63 methods of rebuilding that figure is handled, the result that consecutive frame difference analysis 62 is obtained and the background of carrying out are rebuild 63 results that obtain and are carried out gray processing and smoothing processing 65, after taking above-mentioned image pre-service, use Wavelet Edge Detection 67 methods, the proper vector similarity is calculated 69, calculate the value of Euclidean distance D 71, this is the analysis data of the vehicle front side body structure figure that obtains on the location parameter basis of reality shooting, need be according to the location parameter x of vehicle template, y, α carries out location parameter and proofreaies and correct the judgement that just can be used for after 74 vehicle, according to brand recognition 70 results, database vehicle template parameter 73 and location parameter x, y, α revises that result after 72 analyzes and vehicle judges 74, makes the accuracy rate of judged result of vehicle model specification 75 higher like this.
Fig. 6 is true and false vehicle discriminant analysis journey process flow diagram, by information processing platform system 14 enter call database information 78 programs after, the be called driving board model of going forward side by side of the number-plate number 76 information and database information of vehicles 77 conforms to and 79 judges, carry out the car plate model and conform to and 79 judge according to calling database information 78 and the vehicle model specification of having judged 75, if car plate and vehicle off-gauge then be judged to be false car or false car plate 80,, calls by car plate the identical car plate type information in the database monitoring data 81 in the regulation time limit if conforming to the vehicle specification, have or not similar vehicle 82 to judge, if do not have similar vehicle in the time limit in regulation then be judged to be true car 83, if similar vehicle is arranged then entry time difference analysis 84 programs, time of occurrence difference to each similar vehicle and vehicle to be determined is calculated, determine that carrying out the region road differential again after mistiming that similar vehicle occurs analyses 85, determine the minimum distance that may travel between the similar vehicle according to the geographic position that monitors shooting and regional landform road, carry out speed of a motor vehicle calculating 86 according to minimum distance and mistiming, compare the speed of a motor vehicle of determining this vehicle with the max. speed of the zones of different road of setting and exceed standard not 87, if exceed standard then be judged to be false car or false car plate 80, if do not exceed standard then be judged to be true car 83.
Fig. 7 is the RGB spectroscopic measurements and compares calculation flow chart, vehicle background light 88 scioptics optically focused 89 carry out by prism RGB beam split 90 background optical being decomposed into ruddiness R behind the optically focused, three kinds of basic monochromatic light of green glow G and blue light B, after receiving 91 by RGB monochromatic light CCD, three kinds of monochromatic light are transformed into RGB analog electrical signal 92, the waveform of this analog electrical signal and amplitude and vehicle background light intensity are proportional, analog electrical signal is by A/D conversion output RGB digital signal 93, carry out the synthetic model analysis by RGB digital component to background light, system can calculate the RGB digital component 95 of three kinds of monochromatic light intensity Cd of RGB respectively, calculate the illuminance digital quantity 96 of the illumination Lx of background light simultaneously, usually, the benchmark colour temperature of the background light that shooting needs is in the 4950K-5600K scope, generally be set at 5250K, reference light illumination is 1000Lx, and the gentle illuminance numerical value of these reference color can be set as required.
The RGB digital component 94 of benchmark colour temperature and the RGB digital component 95 of measured vehicle background light are carried out relatively 98 calculating of benchmark light intensity, if the RGB digital component 94 of benchmark colour temperature is less than the RGB digital component 95 of measured vehicle background light then do not need to carry out light filling, program is returned RGB digital component 95 and is proceeded comparison, if the RGB digital component 94 of benchmark colour temperature is greater than the RGB digital component 95 of measured vehicle background light, then respectively to ruddiness R, green glow G and blue light B calculate RGB difference component 100, the ratio of this RGB difference component 100 is exactly the color separation light filling amount ratio that need carry out light filling, calculate the controlled LED constant current of RGB ratio 102 respectively with this, the ratio of RGB difference component 100 is not equal to the controlled LED constant current of RGB ratio 102, and this luminous parameters with three kinds of LED of RGB is relevant.
The illuminance digital quantity 96 of reference light illumination digital quantity 97 and measured vehicle background light is carried out relatively 99 calculating of benchmark illumination, if reference light illumination digital quantity 97 is less than the illuminance digital quantity 96 of measured vehicle background light then do not need to carry out light filling, program back light illumination digital quantity 96 is proceeded comparison, if reference light illumination digital quantity 97 is greater than the illuminance digital quantity 96 of measured vehicle background light, then calculate illuminance measures of dispersion 101, size according to the illuminance measures of dispersion starts infrared light filling 103 judgements, if the illuminance measures of dispersion is greater than preset value, a little less than illustrating that vehicle background light too, normally night or seriously the moon greasy weather gas, at this moment need to carry out infrared light filling, then calculate the controlled LED steady current 105 of RL, if the illuminance measures of dispersion is less than preset value, then enter and calculate controlled LED steady current 104 computation processes of RGB, calculate ruddiness R respectively according to illuminance measures of dispersion and the controlled LED constant current of RGB ratio, the constant current value of green glow G and the controlled LED of blue light B, be nonlinear proportional relation between controlled LED steady current and the rgb light intensity, need carry out certain test and determine.

Claims (6)

1, a kind of car plate vehicle real-time image monitoring and genuine-fake identification system of automatic color regulating light filling, it is characterized in that being divided into front end picture pick-up device and information processing centre two parts, adopt the public/private communication network to carry out data transmission between front end picture pick-up device and the information processing centre, the front end picture pick-up device comprises RGB toning light filling illuminating lamp, monitor and use the colored digital video camera, the RGB spectroscope, illuminating lamp and shooting control circuit, colour temperature is analyzed comparing unit, graphical analysis and algorithm logic, the front end data storehouse, A/D conversion and image acquisition, cpu central processing unit and digital compression hop, information processing centre comprises the Web private server, document data base, information processing platform system and Compress softwares and communication interface.
2, the car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling according to claim 1, it is characterized in that RGB toning light filling illuminating lamp is by redness, green, blue and infrared four LED power models constitute, the RGB spectroscope is measured the colour temperature and the illumination of the background light of vehicle, with image acquisition circuit the background light conversion of signals of measuring is become digital quantity by the A/D conversion, analyzing comparing unit by colour temperature compares the standard color temperature stored in the background light digital quantity measured and the front end data storehouse and illumination and calculates the parameter request that need carry out light filling, control the working current of four kinds of LED power models of RGB toning light filling illuminating lamp respectively by illuminating lamp and shooting control circuit, produce the compensation light of different-colour and illumination, night or rainy weather are opened the infrared illumination lamp and are carried out infrared light filling.
3, the car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling according to claim 1, it is characterized in that two supervision convert digital quantity by the A/D conversion with image acquisition with the vehicle image of colored digital video camera production, instruction according to cpu central processing unit, carry out the image pre-service by graphical analysis and arithmetic logic unit, image segmentation and identification, with the number-plate number of confirming, elementary parameters such as vehicle trade mark and vehicle specification are carried out digital compression transmission back and are carried out detailed comparative analysis and processing by the information vehicle platform system that the public/private communication network sends to information processing centre, and true and false car plate and vehicle are screened.
4, the car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling according to claim 1, it is characterized in that vehicle identification comprised car plate is discerned, brand mark identification and three parts of vehicle structure profile identification, car plate and trade mark are adopted framing, Character segmentation, the sane Hausdorff distance calculating method of Canny operator edge detection and self-adaptation carries out Recognition of License Plate Characters and brand recognition, to improve the discrimination of characters on license plate, two location parameter x that supervision is being set with the colored digital video camera about employing, the vehicle structure image of leaning to one side before two is set respectively in the scope of y and α, by image pre-service and Wavelet Edge Detection, the proper vector similarity is calculated and is carried out location parameter with the vehicle template data of database and proofread and correct back and relatively, determines the brand and the specifications and models of vehicle.
5, the car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling according to claim 1, it is characterized in that the identification of true and false vehicle and car plate is by comparing with database vehicle template information, determine whether car plate conforms to the vehicle specification model, if be not inconsistent then be judged to be false car plate or false car, when vehicle conforms to car plate, then the vehicle data of producing according to each surveillance of a specified time in the zone compares, if do not have same vehicle then be judged to be the Zhen Chezhen car plate, if the same vehicle monitor record is arranged, then for twice or repeatedly record carries out the mistiming and the region road differential is analysed, calculate the rationality of same vehicle, judge whether this vehicle is that a plurality of same model vehicles of car plate use in the time gap of different time and the appearance of different regions.
6, the car plate vehicle real-time image monitoring and the genuine-fake identification system of automatic color regulating light filling according to claim 5, it is characterized in that database vehicle template record adopt identical about two supervision colored digital video cameras, front upper side at vehicle carries out the vehicle image setting, in the assigned position parameter x, two vehicle front side body structure images about producing respectively under y and the α numerical value prerequisite, car plate and brand mark image, adopt Canny operator edge detection and the sane Hausdorff distance calculating method of self-adaptation to determine the image digitization amount parameter value of car plate and brand mark, by the image pre-service, Wavelet Edge Detection and proper vector similarity are calculated and are determined vehicle structure image digitization amount parameter, these car plates with vehicle, the digital quantity parameter of brand mark and vehicle structure is stored in as the template parameter of this vehicle in the document data base of information processing centre as the vehicle template parameter, is used for monitoring vehicle is compared and judges.
CNB2006100684302A 2006-08-23 2006-08-23 Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate Active CN100449576C (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CNB2006100684302A CN100449576C (en) 2006-08-23 2006-08-23 Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CNB2006100684302A CN100449576C (en) 2006-08-23 2006-08-23 Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate

Publications (2)

Publication Number Publication Date
CN1920896A true CN1920896A (en) 2007-02-28
CN100449576C CN100449576C (en) 2009-01-07

Family

ID=37778616

Family Applications (1)

Application Number Title Priority Date Filing Date
CNB2006100684302A Active CN100449576C (en) 2006-08-23 2006-08-23 Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate

Country Status (1)

Country Link
CN (1) CN100449576C (en)

Cited By (25)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101571999A (en) * 2009-06-08 2009-11-04 上海高德威智能交通系统有限公司 Infrared shooting device and traffic information system
CN101635093A (en) * 2009-05-22 2010-01-27 上海宝康电子控制工程有限公司 Identification system and method of unlicensed motor vehicles
CN102053563A (en) * 2010-12-27 2011-05-11 刘宇 Flight training data acquisition and quality evaluation system of analog machine
CN102436746A (en) * 2011-09-29 2012-05-02 杭州海康威视系统技术有限公司 High efficient and real-time analytical method of vehicle fake license suspicion and apparatus thereof
CN102509458A (en) * 2011-10-19 2012-06-20 杭州华三通信技术有限公司 Fake-licensed vehicle identifying method and device
CN102521986A (en) * 2011-12-05 2012-06-27 沈阳聚德视频技术有限公司 Automatic detection system for fake plate vehicle and control method for automatic detection system
CN102881169A (en) * 2012-09-26 2013-01-16 青岛海信网络科技股份有限公司 Fake-licensed car detection method
CN102956107A (en) * 2012-11-16 2013-03-06 广州市华标科技发展有限公司 Parking lot vehicle parking information acquisition and false license, slip license and illegal vehicle recognition system and method
CN103593880A (en) * 2013-09-29 2014-02-19 西安祥泰软件设备系统有限责任公司 Method and device for paying vehicle parking fee
CN103646545A (en) * 2013-11-15 2014-03-19 天津天地伟业数码科技有限公司 Infrared bayonet camera snapshot realization method
CN104301702A (en) * 2014-10-31 2015-01-21 杭州海康威视数字技术股份有限公司 Color temperature value processing method and device and camera capable of adjusting color temperature
CN104408931A (en) * 2014-10-29 2015-03-11 合肥指南针电子科技有限责任公司 Incomplete sign license plate identification system and method
CN104658264A (en) * 2015-01-27 2015-05-27 武汉烽火众智数字技术有限责任公司 Vehicle verification method and system based on ETC and video
CN104954658A (en) * 2014-03-25 2015-09-30 联想(北京)有限公司 Electric device and data processing method
CN105931468A (en) * 2016-07-06 2016-09-07 成都市红亿科技有限公司 Method for automatically regulating and controlling external fill light for license plate identification
CN106488092A (en) * 2015-09-01 2017-03-08 德尔福技术有限公司 Integrated video camera, ambient light detect and rain sensor assembly
CN107393309A (en) * 2017-08-31 2017-11-24 江苏开放大学 A kind of electronic police system
CN110110718A (en) * 2019-03-20 2019-08-09 合肥名德光电科技股份有限公司 A kind of artificial intelligence image processing apparatus
CN110611774A (en) * 2019-09-20 2019-12-24 深圳市梦网百科信息技术有限公司 Illumination compensation method and system for road surface monitoring video
CN111935867A (en) * 2020-08-11 2020-11-13 豪尔赛科技集团股份有限公司 Illumination monitoring system with color correction function
CN113409502A (en) * 2021-06-28 2021-09-17 杭州海康威视数字技术股份有限公司 Method, device and system for controlling vehicle to pass and electronic equipment
CN114354635A (en) * 2022-01-18 2022-04-15 深圳市乐惠光电科技有限公司 Pier crack detection system and method for improving illumination light source
CN114827485A (en) * 2022-04-25 2022-07-29 智道网联科技(北京)有限公司 Light supplementing method and system for vehicle camera
CN117523505A (en) * 2023-10-30 2024-02-06 深圳市大道至简信息技术有限公司 Method for judging real vehicle based on video
CN117746111A (en) * 2023-12-11 2024-03-22 华南理工大学 Method for intelligently judging black smoke grade of diesel vehicle through monitoring images of annular inspection line

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107170239A (en) * 2017-06-30 2017-09-15 广东工业大学 A kind of target vehicle follows the trail of grasp shoot method and device

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1420041A (en) * 2001-11-21 2003-05-28 北京汉王科技有限公司 Embedded integrative vehicle licensing plate distinguishing apparatus
GB2395595B (en) * 2002-11-14 2005-01-05 Nathan Mendel Rau Automated license plate recognition system for use in law enforcement vehicles
CN1553757A (en) * 2003-05-27 2004-12-08 ������ҵ�ɷ����޹�˾ Flashing light controller and operatnig method thereof
CN1725266A (en) * 2004-07-21 2006-01-25 上海高德威智能交通系统有限公司 Vehicle intelligent monitoring recording syste mand method based on video trigger and speed measuring
CN1797496A (en) * 2004-12-30 2006-07-05 汉王科技股份有限公司 Movable type equipment for monitoring vehicles

Cited By (35)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101635093A (en) * 2009-05-22 2010-01-27 上海宝康电子控制工程有限公司 Identification system and method of unlicensed motor vehicles
CN101571999A (en) * 2009-06-08 2009-11-04 上海高德威智能交通系统有限公司 Infrared shooting device and traffic information system
CN102053563A (en) * 2010-12-27 2011-05-11 刘宇 Flight training data acquisition and quality evaluation system of analog machine
CN102436746B (en) * 2011-09-29 2014-10-22 杭州海康威视系统技术有限公司 High efficient and real-time analytical method of vehicle fake license suspicion and apparatus thereof
CN102436746A (en) * 2011-09-29 2012-05-02 杭州海康威视系统技术有限公司 High efficient and real-time analytical method of vehicle fake license suspicion and apparatus thereof
CN102509458A (en) * 2011-10-19 2012-06-20 杭州华三通信技术有限公司 Fake-licensed vehicle identifying method and device
CN102521986A (en) * 2011-12-05 2012-06-27 沈阳聚德视频技术有限公司 Automatic detection system for fake plate vehicle and control method for automatic detection system
CN102881169A (en) * 2012-09-26 2013-01-16 青岛海信网络科技股份有限公司 Fake-licensed car detection method
CN102881169B (en) * 2012-09-26 2015-06-03 青岛海信网络科技股份有限公司 Fake-licensed car detection method
CN102956107B (en) * 2012-11-16 2014-07-16 广州市华标科技发展有限公司 Parking lot vehicle parking information acquisition and false license, slip license and illegal vehicle recognition system and method
CN102956107A (en) * 2012-11-16 2013-03-06 广州市华标科技发展有限公司 Parking lot vehicle parking information acquisition and false license, slip license and illegal vehicle recognition system and method
CN103593880A (en) * 2013-09-29 2014-02-19 西安祥泰软件设备系统有限责任公司 Method and device for paying vehicle parking fee
CN103646545A (en) * 2013-11-15 2014-03-19 天津天地伟业数码科技有限公司 Infrared bayonet camera snapshot realization method
CN104954658A (en) * 2014-03-25 2015-09-30 联想(北京)有限公司 Electric device and data processing method
CN104954658B (en) * 2014-03-25 2019-10-29 联想(北京)有限公司 A kind of electronic equipment and data processing method
CN104408931A (en) * 2014-10-29 2015-03-11 合肥指南针电子科技有限责任公司 Incomplete sign license plate identification system and method
CN104301702A (en) * 2014-10-31 2015-01-21 杭州海康威视数字技术股份有限公司 Color temperature value processing method and device and camera capable of adjusting color temperature
CN104658264A (en) * 2015-01-27 2015-05-27 武汉烽火众智数字技术有限责任公司 Vehicle verification method and system based on ETC and video
CN106488092B (en) * 2015-09-01 2021-01-29 安波福技术有限公司 Integrated camera, ambient light detection and rain sensor assembly
CN106488092A (en) * 2015-09-01 2017-03-08 德尔福技术有限公司 Integrated video camera, ambient light detect and rain sensor assembly
CN105931468B (en) * 2016-07-06 2018-09-11 成都市红亿科技有限公司 A kind of method for automatically regulating of the external light compensating lamp of Car license recognition
CN105931468A (en) * 2016-07-06 2016-09-07 成都市红亿科技有限公司 Method for automatically regulating and controlling external fill light for license plate identification
CN107393309A (en) * 2017-08-31 2017-11-24 江苏开放大学 A kind of electronic police system
CN110110718A (en) * 2019-03-20 2019-08-09 合肥名德光电科技股份有限公司 A kind of artificial intelligence image processing apparatus
CN110110718B (en) * 2019-03-20 2022-11-22 安徽名德智能科技有限公司 Artificial intelligence image processing device
CN110611774A (en) * 2019-09-20 2019-12-24 深圳市梦网百科信息技术有限公司 Illumination compensation method and system for road surface monitoring video
CN110611774B (en) * 2019-09-20 2021-07-16 深圳市梦网视讯有限公司 Illumination compensation method and system for road surface monitoring video
CN111935867A (en) * 2020-08-11 2020-11-13 豪尔赛科技集团股份有限公司 Illumination monitoring system with color correction function
CN111935867B (en) * 2020-08-11 2022-07-26 豪尔赛科技集团股份有限公司 Illumination monitoring system with color correction function
CN113409502A (en) * 2021-06-28 2021-09-17 杭州海康威视数字技术股份有限公司 Method, device and system for controlling vehicle to pass and electronic equipment
CN113409502B (en) * 2021-06-28 2023-03-24 杭州海康威视数字技术股份有限公司 Method, device and system for controlling vehicle to pass and electronic equipment
CN114354635A (en) * 2022-01-18 2022-04-15 深圳市乐惠光电科技有限公司 Pier crack detection system and method for improving illumination light source
CN114827485A (en) * 2022-04-25 2022-07-29 智道网联科技(北京)有限公司 Light supplementing method and system for vehicle camera
CN117523505A (en) * 2023-10-30 2024-02-06 深圳市大道至简信息技术有限公司 Method for judging real vehicle based on video
CN117746111A (en) * 2023-12-11 2024-03-22 华南理工大学 Method for intelligently judging black smoke grade of diesel vehicle through monitoring images of annular inspection line

Also Published As

Publication number Publication date
CN100449576C (en) 2009-01-07

Similar Documents

Publication Publication Date Title
CN100449576C (en) Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate
CN110660222B (en) Intelligent environment-friendly electronic snapshot system for black-smoke road vehicle
US9558412B2 (en) Vehicle exterior environment recognition device
CN109299674B (en) Tunnel illegal lane change detection method based on car lamp
CN103093249B (en) A kind of taxi identification method based on HD video and system
CN106778534B (en) Method for identifying ambient light during vehicle running
CN1725266A (en) Vehicle intelligent monitoring recording syste mand method based on video trigger and speed measuring
CN103808723A (en) Exhaust gas blackness automatic detection device for diesel vehicles
CN1448886A (en) Apparatus and method for measuring vehicle queue length
CN101030256A (en) Method and apparatus for cutting vehicle image
Rojas et al. Vehicle detection in color images
CN112528861A (en) Foreign matter detection method and device applied to track bed in railway tunnel
JPH08233525A (en) Vehicle-type discriminating apparatus
CN113380069A (en) Street lamp-based roadside parking system and method thereof
JP2004086417A (en) Method and device for detecting pedestrian on zebra crossing
CN106340031A (en) Method and device for detecting moving object
CN111724605A (en) Black smoke vehicle monitoring system and monitoring method thereof
CN107066929B (en) Hierarchical recognition method for parking events of expressway tunnel integrating multiple characteristics
CN205788764U (en) The automatic evidence-collecting device of steering indicating light is not opened in a kind of vehicle modified line
CN106128112B (en) Night bayonet vehicle identifies grasp shoot method
US20230401875A1 (en) Method for recognizing illumination state of traffic lights, electronic device, and storage medium
JPH11353580A (en) Method and device for discriminating kind of vehicle at night
CN112985852A (en) Urban rail vehicle detection system
JPH0954891A (en) Method for discriminating vehicle type by image processing and device therefor
JPH0981758A (en) Vehicle detecting device

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant
EE01 Entry into force of recordation of patent licensing contract

Assignee: Shandong Kangwei Communication Technology Co., Ltd.

Assignor: Wang Yasheng

Contract fulfillment period: 2008.12.1 to 2016.8.22

Contract record no.: 2008370000206

Denomination of invention: Real-time image monitoring and genuine-fake identification system for vehicle with automatic color regulating and light filling license plate

Granted publication date: 20090107

License type: Exclusive license

Record date: 20081218

LIC Patent licence contract for exploitation submitted for record

Free format text: EXCLUSIVE LICENSE; TIME LIMIT OF IMPLEMENTING CONTACT: 2008.12.1 TO 2016.8.22; CHANGE OF CONTRACT

Name of requester: SHANDONG KANGWEI COMMUNICATION TECHNOLOGY CO., LTD

Effective date: 20081218

C56 Change in the name or address of the patentee

Owner name: SHANDONG KANGWEI COMMUNICATION TECHNOLOGY CO., LTD

Free format text: FORMER NAME: WANG YASHENG

CP03 Change of name, title or address

Address after: Shandong province Ji'nan City hi tech Development Zone, road 1, Qilu Software Park F1 block A203

Patentee after: Shandong Kangwei Communication Technology Co., Ltd.

Address before: Room 526, main building, South Mountain, No. 73, No. ten, Lixia Road, Lixia District, Shandong, Ji'nan

Patentee before: Wang Yasheng

C56 Change in the name or address of the patentee
CP01 Change in the name or title of a patent holder

Address after: 250101 Shandong city of Ji'nan province high tech Zone Shun Road No. 1, Qilu Software Park F1 block A203

Patentee after: Shandong Conwell Communication Technology Co., Ltd.

Address before: 250101 Shandong city of Ji'nan province high tech Zone Shun Road No. 1, Qilu Software Park F1 block A203

Patentee before: Shandong Kangwei Communication Technology Co., Ltd.

CP03 Change of name, title or address
CP03 Change of name, title or address

Address after: 250101 A203, block F, entrepreneurship square, Qilu Software Park, No.1 Shunhua Road, high tech Development Zone, Jinan City, Shandong Province

Patentee after: Conway Communication Technology Co., Ltd

Address before: 250101 A203, block F1, Qilu Software Park, No.1 Shunhua Road, high tech Zone, Jinan City, Shandong Province

Patentee before: SHANDONG KANGWEI COMMUNICATION TECHNOLOGY Co.,Ltd.