CN101635091B - Device for detecting vehicle and identifying color - Google Patents
Device for detecting vehicle and identifying color Download PDFInfo
- Publication number
- CN101635091B CN101635091B CN2008100408621A CN200810040862A CN101635091B CN 101635091 B CN101635091 B CN 101635091B CN 2008100408621 A CN2008100408621 A CN 2008100408621A CN 200810040862 A CN200810040862 A CN 200810040862A CN 101635091 B CN101635091 B CN 101635091B
- Authority
- CN
- China
- Prior art keywords
- color
- vehicle
- module
- texture
- video
- 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
- Image Analysis (AREA)
- Image Processing (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a device for detecting a vehicle and identifying a color, which can automatically detect the vehicle comprehensively and effectively identify the vehicle color in any environment. The device adopts the technical scheme that: the device comprises a video capturing module for acquiring a video image containing the vehicle, a texture detecting tag which is arranged on a road and is provided with a texture pattern structure to be distinguished from the background of the road, a color marking board which is provided with a plurality of colors and is arranged within the shooting visual field of an image acquisition module, and a control module, wherein the control module further comprises a texture detecting unit for detecting whether the covered area of the texture detecting tag exceeds a preset value, and a color identifying unit which uses the color of the color marking board in the acquired video image as a reference to obtain the deviation degree of the color affected by the environment, calculates the constancy restraint in the current environmental color, and identifies the vehicle color in the video image by using the restraint. The device for detecting the vehicle and identifying the color is applied to identification and detection of the vehicle.
Description
Technical field
The present invention relates to a kind of vehicle detection apparatus, the device that relates in particular to a kind of vehicle detection and vehicle color is discerned.
Background technology
Vehicle Automatic Measurement Technique in the vehicle automatic management is one of the most key technology, and this will determine whether follow-up management can effectively handle related data.Preferably under the situation, the video capture module during vehicle detects automatically can obtain video image clearly at light, and this just can successfully be embedded into common car detection recognition system in the vehicle automatic management system.
But traditional vehicle identification detects following deficiency is arranged: the algorithm that (1) only relies on car plate identification, can't effectively detect unlicensed or dirty board vehicle, will cause omission, vehicle management system can not effectively be worked; (2) though traditional motion analysis algorithm can fine detection vehicle, motion analysis is extremely responsive to environment, blocked influence big, and for can not well working under the situation at inclement weather or night; (3) motion analysis generally speaking, object classification scheduling algorithm complexity, calculated amount is big, and resource consumption is many, for making up the threat that real-time vehicle management system has reality.
Vehicle color is one of important external appearance characteristic in the vehicle management system, and the detection and Identification of vehicle color at present are also lower, greatly about about 75%.Because light changes, shade disturbs and the effective color region of car body extracts, make effective identification of vehicle color become the very problem of difficulty.
To sum up, the problem that faces at present mainly contains: (1) in any environment efficiently vehicle do not have to leak automatically and detect; (2) vehicle color identification, the technical barrier that the former faces is the long-time stability of distinguishing and moving and the high efficiency of vehicle, the technical barrier that the latter faces is that the color constancy under any environment is estimated.
Summary of the invention
The objective of the invention is to address the above problem, a kind of vehicle detection and color recognition device are provided, can both not have in any environment and leak the self check vehicle and vehicle color is effectively discerned.
Technical scheme of the present invention is: the present invention has disclosed a kind of vehicle detection and color recognition device, comprising:
Video capture module is used to gather the video image that comprises vehicle;
The texture certification mark is installed on the road surface, and the line case structure with texture is to be different from the road surface background;
The color calibration board which is provided with multiple color, is installed in the shooting visual field of this video capture module;
Control module further comprises:
The texture detecting unit detects the area that this texture certification mark is blocked and whether exceeds preset value;
The color recognition unit is reference with the color of the color calibration board in the video image that collects, and obtains color irrelevance affected by environment, calculates the shape constancy constraint of current environment color and utilizes this constraint to discern vehicle color in this video image.
Above-mentioned vehicle detection and color recognition device, wherein, this texture certification mark and this color calibration board combine, and form the colour band line of stumbling, the texture that this colour band is stumbled on the line is the alternate color lump of multiple color, perpendicular to the vehicle heading setting and with the plane parallel of this video capture module.
Above-mentioned vehicle detection and color recognition device, wherein, this video capture module is a video camera.
Above-mentioned vehicle detection and color recognition device, wherein, this device also comprises:
Flash modules is assisted this video capture module work;
This control module also comprises:
The flashlamp control module starts this flash modules under the situation of insufficient light.
Above-mentioned vehicle detection and color recognition device, wherein, this device also comprises:
Virtual detection mark and this texture certification mark are arranged on same place;
Control module also comprises:
The virtual detection unit detects this virtual detection and marks whether to be triggered.
Above-mentioned vehicle detection and color recognition device, wherein, this control module also comprises:
The car light detecting unit judges as the auxiliary detection object whether vehicle enters in the shooting visual field of this video capture module with car light.
Above-mentioned vehicle detection and color recognition device, wherein, this device also comprises:
The infrared video capture module is used for comprising with the infrared mode collection video image of vehicle;
Infrared supplementary lighting module is assisted this infrared video capture module work;
This control module also comprises:
Infrared acquisition controlling unit starts this infrared video capture module during less than a threshold value in light illumination;
Infrared light filling control module is controlled the running of this infrared supplementary lighting module.
The present invention contrasts prior art following beneficial effect: the present invention can effectively not have the self check of leakage vehicle by colour band, and utilizes color constancy that vehicle color is carried out effective recognition by the color calibration board.And scene physical constraint then and there is subjected to will starting the auxiliary detection mode automatically, for example virtual stumble line or car light detection etc. after ignorance factor is destroyed.For the mode of traditional video capture car plate, detection mode of the present invention is more simple effectively.
Description of drawings
Fig. 1 is the block diagram of first embodiment of vehicle detection of the present invention and color recognition device.
Fig. 2 A is the block diagram of second embodiment of vehicle detection of the present invention and color recognition device.
Fig. 2 B be among second embodiment of vehicle detection of the present invention and color recognition device vehicle through the stumble synoptic diagram of line of colour band.
Fig. 3 is the block diagram of the 3rd embodiment of vehicle detection of the present invention and color recognition device.
Fig. 4 A is the block diagram of the 4th embodiment of vehicle detection of the present invention and color recognition device.
Fig. 4 B be among the 4th embodiment of vehicle detection of the present invention and color recognition device vehicle through the stumble synoptic diagram of line and the virtual line of stumbling of colour band.
Fig. 5 is the block diagram of the 5th embodiment of vehicle detection of the present invention and color recognition device.
Fig. 6 is the block diagram of the 6th embodiment of vehicle detection of the present invention and color recognition device.
Fig. 7 A is the block diagram of the example of vehicle management system of the present invention.
Fig. 7 B is the schematic flow sheet of the vehicle management system of Fig. 7 A.
Embodiment
The invention will be further described below in conjunction with drawings and Examples.
Fig. 1 shows first embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 1, device comprises texture certification mark 10, color calibration board 11, video capture module 12, control module 13, and its middle controller 13 further comprises texture detecting unit 130 and color recognition unit 131.
Texture certification mark 10 is installed on the scene of road surface, is positioned at the shooting visual field of video capture module 12, and the line case structure with remarkable texture for example can be the mark of lattice formation or the mark that literal constitutes to be different from the background on road surface.Color calibration board 11 also is installed in the shooting visual field of video capture module 12, which is provided with multiple color, for example comprises black, red, green, blue, white alternate color lump.Certainly, color calibration board 11 also can have other different forms, as be coated in color on the colour band, it also can be the shades of colour that is coated on the back brake, or one within the vision of video capture module independently scribbles versicolor metal board, its shape can be according to concrete Environment Design, for example square, circular, triangle, bar shaped etc.
Texture detecting unit 130 in the control module 13 judges whether the area that texture certification mark 10 is blocked exceeds preset value.This is owing to vehicle will cover certain, continuous texture certification mark zone by making, and the lasting regular hour, and occupy relatively large area in the entire image, the probability that people or non-regular crowd can not have above-mentioned feature or generation simultaneously is extremely low, if can be lower in conjunction with the whole probable value of particular environment, so this area detection mode of texture detecting unit 130 fully can practical requirement.In texture detecting unit 130, exceed preset value if judge the area that texture certification mark 10 is blocked, vehicle then is described through texture certification mark 10, comprise vehicle on the video image that gathered by video capture module 12 (for example being a video camera) this moment.
On the video image of gathering, also comprised color calibration board 11, color recognition unit 131 serves as with reference to obtaining the irrelevance that color is influenced by prevailing circumstances with the color of the color calibration board 11 in the video image that collects, calculate the shape constancy constraint of current environment color, utilize the color of the vehicle in this constraint identification video image itself.
Under the night situation, also can utilize the flashlamp (not shown in figure 1) to obtain clearly and do identification again in the video image.
Fig. 2 A shows second embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 2 A, device comprises colour band stumble line 20, video capture module 21 and control module 22, and wherein control module 22 further comprises texture detecting unit 220 and color recognition unit 221.
With respect to first embodiment, present embodiment combines color calibration board and texture certification mark, forms the colour band line 20 of stumbling.The texture that colour band is stumbled on the line 20 is the alternate color lump of multiple color, for example be deceive, red, green, blue, white alternate color lump, be installed on the road surface, perpendicular to vehicle heading and with the plane parallel of video capture module 21.Use the many colour bands line 20 of stumbling in the present embodiment.
Please be simultaneously referring to Fig. 2 B, the colour band line 20 of stumbling is provided with black color lump 200, red color lump 201, green color lump 202, blue color lump 203 and white color lump 204, vehicle 23 through colour bands stumble line 20 make can shield portions the colour band line 20 of stumbling.
The color that color recognition unit 221 is stumbled on the line 20 with the colour band in the video image that collects serves as with reference to obtaining the irrelevance that color is influenced by prevailing circumstances, calculate the shape constancy constraint of current environment color, utilize the color of the vehicle in this constraint identification video image itself.
Under the night situation, also can utilize flashlamp (not shown among Fig. 2 A) to obtain clearly and do identification again in the video image.
Fig. 3 shows the 3rd embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 3, device comprises texture certification mark 30, virtual detection mark 31, color calibration board 32, video capture module 33, flash modules 34 and control module 35.Wherein control module 35 further comprises texture detecting unit 350, virtual detection unit 351, color recognition unit 352 and flashlamp control module 353.
Under texture certification mark 30 apparent situations, device operates with following workflow: the texture detecting unit 350 in the control module 35 judges whether the area that texture certification mark 30 is blocked exceeds preset value.Exceed preset value if judge the area that texture certification mark 30 is blocked, vehicle then is described, comprise vehicle on the video image that gathered by video capture module 33 (for example being a video camera) this moment through texture certification mark 30.
On the video image of gathering, also comprised color calibration board 32, color recognition unit 352 serves as with reference to obtaining the irrelevance that color is influenced by prevailing circumstances with the color of the color calibration board 32 in the video image that collects, calculate the shape constancy constraint of current environment color, utilize the color of the vehicle in this constraint identification video image itself.
If insufficient light (for example time-division) between the lights then started flash modules 34 by flashlamp control module 353 and carries out light filling at that time, assist video capture module 33 to obtain distinct image.
Under texture certification mark 30 sightless situations (for example big rainy day or snow sky make texture certification mark 30 be beyond recognition), then judge by virtual detection unit 351 whether virtual detection mark 31 (usually and texture certification mark 30 be arranged on same place) is triggered, if be triggered then start color recognition unit 352.Same, when virtual detection triggers, start flash modules 34 collaborative works by flashlamp control module 353.
Fig. 4 A shows the 4th embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 4 A, device comprises colour band stumble line 40, the virtual line 41 of stumbling, video capture module 42, flash modules 43 and control module 44.Wherein control module 44 further comprises texture detecting unit 440, virtual detection unit 441, color recognition unit 442 and flashlamp control module 443.
Compared to the 3rd above-mentioned embodiment, present embodiment combines color calibration board and texture certification mark, forms the colour band line 40 of stumbling.The texture that colour band is stumbled on the line 40 is the alternate color lump of multiple color, for example be deceive, red, green, blue, white alternate color lump, be installed on the road surface, perpendicular to vehicle heading and with the plane parallel of video capture module 42.Use the many colour bands line 40 of stumbling in the present embodiment.In addition, the virtual line 41 of stumbling in the present embodiment is equivalent to the virtual detection mark 31 among the 3rd embodiment.
Under colour band was stumbled line 40 apparent situations, device was with the running of following workflow: the texture detecting unit 440 in the control module 44 judges whether the area of stumbling colour band line 40 being blocked exceeds preset value.Exceed preset value if judge the colour band area that line 40 is blocked of stumbling, vehicle then is described, comprise vehicle on the video image that gathered by video capture module 42 (for example being a video camera) this moment through the colour band line 40 of stumbling.
Please be simultaneously referring to Fig. 4 B, the colour band line 40 of stumbling is provided with black color lump 400, red color lump 401, green color lump 402, blue color lump 403 and white color lump 404, vehicle 45 through colour bands stumble line 40 make can shield portions the colour band line 40 of stumbling.
The color that color recognition unit 442 is stumbled on the line 40 with the colour band in the video image that collects serves as with reference to obtaining the irrelevance that color is influenced by prevailing circumstances, calculate the shape constancy constraint of current environment color, utilize the color of the vehicle in this constraint identification video image itself.
If insufficient light (for example time-division) between the lights then started flash modules 43 by flashlamp control module 443 and carries out light filling at that time, assist video capture module 42 to obtain distinct image.
Under colour band is stumbled line 40 sightless situations (for example big rainy day or snow angel get the colour band line 40 of stumbling and be beyond recognition), then judge by virtual detection unit 441 whether the virtual line 41 of stumbling (usually and the colour band line 40 of stumbling be arranged on same place) is triggered, if be triggered then start color recognition unit 352.Same, when virtual detection triggers, start flash modules 43 collaborative works by flashlamp control module 443.
Fig. 5 shows the 5th embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 5, device comprises texture certification mark 50, virtual detection mark 51, color calibration board 52, video capture module 53, flash modules 54, infrared video capture module 55, infrared supplementary lighting module 56, control module 57.Wherein control module 57 further comprises texture detecting unit 570, virtual detection unit 571, color recognition unit 572, car light detecting unit 573, infrared video seizure control module 574, infrared light filling control module 575 and flashlamp control module 576.
Under texture certification mark 50 apparent situations, device operates with following workflow: the texture detecting unit 570 in the control module 57 judges whether the area that texture certification mark 50 is blocked exceeds preset value.Exceed preset value if judge the area that texture certification mark 50 is blocked, vehicle then is described, comprise vehicle on the video image that gathered by video capture module 53 (for example being a video camera) this moment through texture certification mark 50.
On the video image of gathering, also comprised color calibration board 52, color recognition unit 572 serves as with reference to obtaining the irrelevance that color is influenced by prevailing circumstances with the color of the color calibration board 52 in the video image that collects, calculate the shape constancy constraint of current environment color, utilize the color of the vehicle in this constraint identification video image itself.
If insufficient light (for example time-division) between the lights then started flash modules 54 by flashlamp control module 576 and carries out light filling at that time, assist video capture module 53 to obtain distinct image.
Under texture certification mark 50 sightless situations (for example big rainy day or snow sky make texture certification mark 50 be beyond recognition), then judge by virtual detection unit 571 whether virtual detection mark 51 (usually and texture certification mark 50 be arranged on same place) is triggered, if be triggered then start color recognition unit 572.Same, when virtual detection triggers, start flash modules 54 collaborative works by flashlamp control module 573.
In addition, present embodiment also has the mode of another kind of infrared triggering: for example in the night scene, infrared video catch control module 574 light illumination during less than predetermined threshold value control infrared video capture module 55 with infrared mode images acquired information, infrared supplementary lighting module 56 collaborative works of simultaneously infrared light filling control module 575 controls.The gray level image that obtains under this infrared mode is sent into and has been judged whether in the texture detecting unit 570 that vehicle passes through.Ask video capture module 53 and flash modules 54 synchronous operations simultaneously, color image shot carries out color identification so that carry out color recognition unit 572.
Present embodiment can also carry out auxiliary detection by car light detecting unit 573, judges for the auxiliary detection object whether vehicle is included in the image with the car light.
Fig. 6 shows the 6th embodiment of vehicle detection of the present invention and color recognition device.See also Fig. 6, device comprises: colour band stumble line 61, the virtual line 62 of stumbling, video capture module 63, flash modules 64, infrared video capture module 65, infrared supplementary lighting module 66 and control module 67.Wherein control module 67 further comprises texture detecting unit 670, virtual detection unit 671, color recognition unit 672, car light detecting unit 673, infrared video seizure control module 674, infrared light filling control module 675 and flashlamp control module 676.
Compared to the 5th embodiment shown in Figure 5, present embodiment combines color calibration board and texture certification mark, forms the colour band line 61 of stumbling.The texture that colour band is stumbled on the line 61 is the alternate color lump of multiple color, for example be deceive, red, green, blue, white alternate color lump, be installed on the road surface, perpendicular to vehicle heading and with the plane parallel of video capture module 63.Use the many colour bands line 62 of stumbling in the present embodiment.In addition, the virtual line 62 of stumbling in the present embodiment is equivalent to the virtual detection mark 51 among the 3rd embodiment.
Under colour band was stumbled line 61 apparent situations, device was with the running of following workflow: the texture detecting unit 670 in the control module 67 judges whether the area of stumbling colour band line 61 being blocked exceeds preset value.Exceed preset value if judge the colour band area that line 61 is blocked of stumbling, vehicle then is described, comprise vehicle on the video image that gathered by video capture module 63 (for example being a video camera) this moment through the colour band line 61 of stumbling.
If insufficient light (for example time-division) between the lights then started flash modules 64 by flashlamp control module 676 and carries out light filling at that time, assist video capture module 63 to obtain distinct image.
Under colour band is stumbled line 61 sightless situations (for example big rainy day or snow angel get the colour band line 61 of stumbling and be beyond recognition), then judge by virtual detection unit 671 whether the virtual line 62 of stumbling (usually and the colour band line 61 of stumbling be arranged on same place) is triggered, if be triggered then start color recognition unit 672.Same, when virtual detection triggers, start flash modules 64 collaborative works by flashlamp control module 673.
In addition, present embodiment also has the mode of another kind of infrared triggering: for example in the night scene, infrared video catch control module 674 light illumination during less than predetermined threshold value control infrared video capture module 65 with infrared mode images acquired information, infrared supplementary lighting module 66 collaborative works of simultaneously infrared light filling control module 675 controls.The gray level image that obtains under this infrared mode is sent into and has been judged whether in the texture detecting unit 670 that vehicle passes through.Ask video capture module 63 and flash modules 64 synchronous operations simultaneously, color image shot carries out color identification so that carry out color recognition unit 672.
Present embodiment can also carry out auxiliary detection by car light detecting unit 673, judges for the auxiliary detection object whether vehicle is included in the image with the car light.
Fig. 7 A and Fig. 7 B show the example of vehicle management system of the present invention jointly.Please be simultaneously referring to Fig. 7 A and Fig. 7 B, in system, the scene configuration of video induction comprises vehicle, road surface and the colour band line of stumbling, good in order to obtain, the video image that has the colour band texture is prepared.Video acquiring module comprises video camera, light filling lamp, capture card and computing machine, and the simulating signal by the capture card acquisition camera is in the internal memory of computing machine.Video processing module carries out the check of color and car plate identification, the detection record vehicle, and the identification vehicle number outputs results to higher level's vehicle management module, if recognition failures then feed back to next module.The flashlamp supplementary lighting module comprises flashlamp control module and hardware flashlamp, and the flashlamp control module starts flashlamp according to higher level's feedback result, good bias light is provided for the video induction module.
The System Operation process is as follows: at first carry out system initialization, for example colour band is stumbled the line detection as acquiescence selection, image size, formatting, the loading of colour band Standard Colors, the loading of various algorithms and initialization etc.
Secondly, the scene configuration that has colour band.By being installed on carriageway surfacing of exercising and the road surface, colour band constitutes. and its major function is a stack colour band information in live video stream, for the later stage provides information source based on the Algorithm Analysis of colour band information.
Then, enter the running of video acquiring module.In this example, video acquiring module comprises two submodules: the A module comprises that natural light LED light filling lamp and video camera and B module comprise infrared LED light filling lamp and thermal camera.Two modules of A, B can independently be worked, and need not an other side, also can be with cooperation way work.The A module can be arrested the vehicle pictures that has license board information night, if cooperate flashlamp can obtain the colour picture of whole vehicle.This function can be used to discern vehicle license, the information that car plate color, vehicle color etc. are abundant more.But not having under the sightless situation of secondary light source and colour band, as the heavy snow at night, heavy rain under the condition such as colour band is contaminated, may make the A module can't obtain to have the effective video of colour band.The B module adopts the infrared light filling of LED, and the illumination of sensing light automaticallyes switch between white light pattern and low-light (level) infrared mode automatically, therefore can obtain the effective video under any illumination condition.Greater than certain illumination threshold values, the B module is the normal white light mode of operation, is lower than this value, and the B module is operated in infrared mode, acquisition make gray level image.
The vehicle detection of reruning module, corresponding with video acquiring module, also be divided into A module and B module.The video content of A resume module video acquiring module A, the content of B resume module video acquiring module B.
Modules A mainly comprises the clear detection sub-module of colour band, based on the vehicle detection submodule of colour band, based on the vehicle detection submodule of the virtual line of stumbling, based on the vehicle detection submodule and the communicator module of car light; Module B also comprises the luminance detection module except that comprising the submodule that A comprises, whether whether the picture that detects current collection gray level image, promptly be operated under the infrared mode.Modules A and B have separately independently logic working system, can be operated under primary module and the supplementary module pattern according to different any one modules that requires.The B module is mainly used in the vehicle detection under any external condition, but do not have the color video information at night, make the colouring information of car plate and car body to obtain, the B module can be sent the request of assisting: request color video and flashlamp are synchronous, if the A module exists then response request, require flashlamp synchronous simultaneously.The B module is selected to detect based on colour band according to illuminance information and the clear information of colour band automatically, or the virtual line detection of stumbling, or car light detects.These three submodules are relatively independent, both can work in series, also can concurrent working, and one of them or several assistance authentication modules as main detection module, final assistance information and vehicle detecting information will output to next module.
The A module is mainly used in the vehicle detection under certain constraint condition, and as colour band as seen, perhaps surrounding environment light is better, but under the light filling of flashlamp, it can obtain color video stream preferably.If colour band detects failure or other functions are enabled failure, it can be assisted in order to request module B, if B replys and ask success, then modules A current vehicles measuring ability will temporarily be replaced by module B.When illumination is higher than a certain threshold values, module B will stop this agency service.Modules A will be according to colour band information, and the ambient light information or the solicited message that extract in the video flowing are selected to detect based on colour band automatically, or the virtual line detection of stumbling, or car light detects.These three submodules are relatively independent, both can work in series, also can concurrent working, and one of them or several assistance authentication modules as main detection module. final assistance information and vehicle detecting information will output to next module.
Message management module (Event Adapter) is used for communicating by letter and synchro control mechanism of Coordination module A, B and flashlamp control module.This message management module makes the cooperation of intermodule become simple, flexible, configurable.This framed structure has extremely strong extensibility, and not only each intermodule both can work alone, cooperation that again can be continuously, but also can add new functional module as required.
At last, carry out vehicle license identification and vehicle color identification.From the processing through assistance information processing module and state detection module of the assistance information of vehicle detection module input and testing result, car plate identification and color identification module will obtain the current detection trigger method, obtain related resource then.For example, if Video Detection information from modules A, and is that colour band triggers, picture is clear, and light is better, and the car plate recognizer will normally start so, has the color calibration that colour band color in the dynamic video of colour band information will be used in the vehicle color recognizer.If Video Detection information is equally from modules A, but be not to be that colour band triggers, but trigger by the local virtual line of stumbling, light a little less than, colour band is fuzzy to lose, and the car plate recognizer still normally starts so, but the colour band information in the dynamic video does not exist, at this moment it is (similar with colour band to rely on backup color calibration board, but be not to be laid on the ground, but other positions in the visual field do not mark in system diagram), if neither available, with only according to the information Recognition vehicle color of picture self.If Video Detection information from module B, detects the assistance information of A simultaneously, the car plate recognizer will start A road video so, or time-delay wait A road video, at this moment, modules A is in blocked state, video is not being performed an analysis, and directly directly handling by car plate and color recognizer.If Video Detection information is equally from module B, and well lighted, colour band is clear, and the car plate recognizer will normally start so, has the color calibration that colour band color in the dynamic video of colour band information will be used in the vehicle color recognizer.If Video Detection information is equally from module B, for the local virtual line of stumbling triggers, and colour band is fuzzy invisible, light is darker, if the assistance information of B exists, then recognizer is waited for the input of video module A, if assistance information does not exist, recognizer directly starts.
Final recognition result is inserted database, the vehicle that record this time detects, car plate, color, time, some information such as direction.
Application scenarios of the present invention is as follows:
One, daytime: (1) is used for vehicle detection; (2) be used for vehicle color comparison and identification; (3) for the vehicle of no car plate or car plate messy can start flashlamp, capture and know picture, file; (4) reduction improves identification accuracy and recognition efficiency to the consumption of computer resource.
Two, evening: after (1) light deepening, the video image through behind the LED light filling will have a tangible reflective colour band; (2) be used for vehicle detection; (3) if doubtful vehicle, but not identification then starts flashlamp, captures clear picture, files; (4) be used for vehicle color comparison and identification; (5) reduction improves identification accuracy and recognition efficiency to the consumption of computer resource.
Three, the rainy day: after (1) light deepening, through the video image behind the LED light filling, if there is not a tangible reflective colour band, then there is ponding (raining) on contaminated or road surface, ground, starts common light source light filling, stroke virtual line of stumbling, detection vehicle; (2) the virtual line of stumbling is triggered, and starts the flashlamp light filling, captures clear picture, and identification is filed; (3) reduction improves identification accuracy and recognition efficiency to the consumption of computer resource.
Four, snow sky: (1) ground is reflective strong, can't obtain a tangible colour band, and the virtual line of stumbling is set; (2) the virtual line of stumbling is triggered, and starts recognizer; (3) reduction improves identification accuracy and recognition efficiency to the consumption of computer resource.
The foregoing description provides to those of ordinary skills and realizes or use of the present invention; those of ordinary skills can be under the situation that does not break away from invention thought of the present invention; the foregoing description is made various modifications or variation; thereby protection scope of the present invention do not limit by the foregoing description, and should be the maximum magnitude that meets the inventive features that claims mention.
Claims (7)
1. vehicle detection and color recognition device comprise:
Video capture module is used to gather the video image that comprises vehicle;
The texture certification mark is installed on the road surface, and the line case structure with texture is to be different from the road surface background;
The color calibration board which is provided with multiple color, is installed in the shooting visual field of this video capture module;
Control module further comprises:
The texture detecting unit detects the area that this texture certification mark is blocked and whether exceeds preset value;
The color recognition unit is reference with the color of the color calibration board in the video image that collects, and obtains color irrelevance affected by environment, calculates the shape constancy constraint of current environment color and utilizes this constraint to discern vehicle color in this video image.
2. vehicle detection according to claim 1 and color recognition device, it is characterized in that, this texture certification mark and this color calibration board combine, form the colour band line of stumbling, the texture that this colour band is stumbled on the line is alternate color lump of multiple color or literal, perpendicular to the vehicle heading setting and with the plane parallel of this video capture module.
3. vehicle detection according to claim 1 and color recognition device is characterized in that, this video capture module is a video camera.
4. vehicle detection according to claim 1 and color recognition device is characterized in that, this device also comprises:
Flash modules is assisted this video capture module work;
This control module also comprises:
The flashlamp control module starts this flash modules under the situation of insufficient light.
5. vehicle detection according to claim 1 and color recognition device is characterized in that, this device also comprises:
Virtual detection mark and this texture certification mark are arranged on same place;
Control module also comprises:
The virtual detection unit detects this virtual detection and marks whether to be triggered.
6. vehicle detection according to claim 1 and color recognition device is characterized in that, this control module also comprises:
The car light detecting unit judges as the auxiliary detection object whether vehicle enters in the shooting visual field of this video capture module with car light.
7. vehicle detection according to claim 1 and color recognition device is characterized in that, this device also comprises:
The infrared video capture module is used for comprising with the infrared mode collection video image of vehicle;
Infrared supplementary lighting module is assisted this infrared video capture module work;
This control module also comprises:
Infrared video is caught control module, starts this infrared video capture module during less than a threshold value in light illumination;
Infrared light filling control module is controlled the running of this infrared supplementary lighting module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2008100408621A CN101635091B (en) | 2008-07-23 | 2008-07-23 | Device for detecting vehicle and identifying color |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2008100408621A CN101635091B (en) | 2008-07-23 | 2008-07-23 | Device for detecting vehicle and identifying color |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101635091A CN101635091A (en) | 2010-01-27 |
CN101635091B true CN101635091B (en) | 2010-12-22 |
Family
ID=41594271
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2008100408621A Expired - Fee Related CN101635091B (en) | 2008-07-23 | 2008-07-23 | Device for detecting vehicle and identifying color |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101635091B (en) |
Families Citing this family (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101807248A (en) * | 2010-03-30 | 2010-08-18 | 上海银晨智能识别科技有限公司 | System for detecting the existence of people in ATM machine video scene and method thereof |
CN102393902A (en) * | 2011-12-01 | 2012-03-28 | 昆山市工业技术研究院有限责任公司 | Vehicle color detection method based on H_S two-dimensional histogram and regional color matching |
CN106250827B (en) * | 2016-07-20 | 2019-06-11 | 浙江宇视科技有限公司 | Vehicle color identification method and device |
CN107729812B (en) * | 2017-09-18 | 2021-06-25 | 南京邮电大学 | Method suitable for recognizing vehicle color in monitoring scene |
JP2019117581A (en) * | 2017-12-27 | 2019-07-18 | トヨタ自動車株式会社 | vehicle |
CN110187703A (en) * | 2019-04-08 | 2019-08-30 | 国以贤智能科技(上海)有限公司 | The method and fork truck of vision guided navigation |
CN116543567B (en) * | 2023-06-29 | 2023-09-29 | 功道(深圳)科技实业有限公司 | License plate discernment all-in-one |
Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1924950A (en) * | 2005-08-30 | 2007-03-07 | 北京允益科数字网络有限公司 | Vehicle characters recognition and measuring system and vehicle information recognition method |
-
2008
- 2008-07-23 CN CN2008100408621A patent/CN101635091B/en not_active Expired - Fee Related
Patent Citations (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1924950A (en) * | 2005-08-30 | 2007-03-07 | 北京允益科数字网络有限公司 | Vehicle characters recognition and measuring system and vehicle information recognition method |
Non-Patent Citations (3)
Title |
---|
JP特开2000-222673A 2000.08.11 |
JP特开2005-285014A 2005.10.13 |
JP特开2008-152627A 2008.07.03 |
Also Published As
Publication number | Publication date |
---|---|
CN101635091A (en) | 2010-01-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101635091B (en) | Device for detecting vehicle and identifying color | |
CN110197589B (en) | Deep learning-based red light violation detection method | |
CN100507970C (en) | Red light overriding detection system and method based on digital video camera | |
JP4035910B2 (en) | Car color discrimination device | |
CN109194873B (en) | Image processing method and device | |
KR101688695B1 (en) | Apparatus and method for recognizing the number of cars and a computer-readable recording program for performing the said method, the recording medium | |
CN103294994A (en) | Apparatus for recognizing face based on environment adaptation | |
CN102156862B (en) | License plate recognition system and license plate recognition method preventing blocking and altering | |
JP6720931B2 (en) | Vehicle surrounding recognition device | |
CN104134352A (en) | Video vehicle characteristic detection system and detection method based on combination of long exposure and short exposure | |
KR101625538B1 (en) | Car Number Recognition system | |
CN107749055A (en) | A kind of fault detection method, system and the device of LED traffic guidances screen | |
CN110730531A (en) | LED street lamp remote light-changing control system | |
CN106469292A (en) | License plate recognition control system and its identification control method | |
CN102568202A (en) | License plate sheltering detection system and detection method | |
KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
US8229170B2 (en) | Method and system for detecting a signal structure from a moving video platform | |
CN108205897A (en) | A kind of car plate analysis and identification system of vehicle | |
KR102473851B1 (en) | Illegal parking management system of parking area for disabled persons | |
CN108810420A (en) | Control method, system and the Car license recognition instrument of light compensating lamp | |
CN202257940U (en) | License plate identification video server | |
CN108710843B (en) | Face detection method and device for attendance checking | |
CN104118352B (en) | Vehicle assisted system, vehicle auxiliary method and vehicle | |
CN105513372B (en) | The anti-capture recognition methods for clapping device vehicle of car plate is installed | |
CN104464287A (en) | Detection system for parking of vehicles at parking places on basis of pattern recognition |
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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20101222 |