CN107256397A - License plate number recognition methods in traffic video image - Google Patents
License plate number recognition methods in traffic video image Download PDFInfo
- Publication number
- CN107256397A CN107256397A CN201710442152.0A CN201710442152A CN107256397A CN 107256397 A CN107256397 A CN 107256397A CN 201710442152 A CN201710442152 A CN 201710442152A CN 107256397 A CN107256397 A CN 107256397A
- Authority
- CN
- China
- Prior art keywords
- license plate
- plate number
- information
- video image
- identified
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
Abstract
In order to improve the Car license recognition accuracy rate to vehicle under different motion speed, while the reduction of identification cost is able to ensure that again, the invention provides the license plate number recognition methods in a kind of traffic video image, including：(1) shoot video image in predetermined shooting area and obtain the geographical location information of the shooting area, the video image includes the traffic video image that pending license plate number is recognized；(2) natural lighting brightness is recorded while video image is shot；(3) obtain and refer to license plate number information；(4) based on the video image for shooting acquisition, by means of the natural lighting brightness and the reference license plate number information, the license plate number in the video image is recognized.
Description
Technical field
The present invention relates to image graphics identification technology field, more particularly, to the car plate in a kind of traffic video image
Number recognition methods.
Background technology
License plate recognition technology occupies vital status in intelligent traffic monitoring system.Processing based on video image
And analysis, the development of intelligent transportation is greatly promoted, is played in traffic surveillance and control system and maintains traffic safety operation
Important function.Intelligent traffic monitoring technology is exactly to carry out IMAQ to traffic scene by front-end image collecting device, to adopting
Vehicle in the image of collection is identified；And traffic scene image is sent to background server, deposited by background server
Storage；Then, the video code flow that background server can be called to store as needed, to obtain transport information.License plate recognition technology
General processing includes Character segmentation and Recognition of License Plate Characters.
Through retrieval, in the prior art usually used algorithm be all directly against vehicle license image enter line character know
Not.For example, Application No. CN201410267812 Chinese invention patent application discloses a kind of Car license recognition based on video
Method, with the automobile video frequency of camera real scene shooting as input, detects moving vehicle and is split, by pretreatment
Target vehicle image afterwards carries out Vertical edge detection, determines the accurate location of license plate area, and License Plate Segmentation is come out, right
License plate image carries out color correction, binaryzation and slant correction processing, by each Character segmentation in oriented license plate area
Out, as single character, feature extraction is carried out to each character, resulting characteristic vector is passed through into good point of training in advance
Class device is classified and using classification results as preliminary recognition result, and defaced characters on license plate is passed through into Prosthetic Hand visual characteristic
Template matching algorithm be recognized, finally give license plate recognition result.
However, these Processing Algorithms used in the prior art are excessively complicated, the quality degree of dependence to image itself is high,
And then add the cost of video monitoring equipment and the data volume of image storage, transmission and processing.All kinds of it is directed in addition, existing
The discrimination of the method for car plate Direct Recognition is not high, especially for the Car license recognition accuracy rate of the faster vehicle of movement velocity
It is relatively low.And the mode that recognition accuracy is improved generally is exactly that this enters one using graphics capability strong server or computer
Step adds the cost of Car license recognition.
The content of the invention
In order to improve the Car license recognition accuracy rate to vehicle under different motion speed, while being able to ensure that identification cost again
Reduction, the invention provides the license plate number recognition methods in a kind of traffic video image, including：
(1) shoot video image in predetermined shooting area and obtain the geographical location information of the shooting area, institute
Stating video image includes the traffic video image that pending license plate number is recognized；
(2) natural lighting brightness is recorded while video image is shot；
(3) obtain and refer to license plate number information；
(4) based on the video image for shooting acquisition, by means of the natural lighting brightness and the reference license plate number information,
Recognize the license plate number in the video image.
Further, the step (3) includes：
(31) vehicle with predetermined vehicle, the video image that is taken is communicated, and obtains its driving information；
(32) according to the geographical location information of the shooting area, determine the predetermined vehicle by the region
Time or moment, it is designated as referring to the moment, and remembers that with this be reference light degree of illuminating with reference to moment corresponding natural lighting brightness；
(33) according to the reference moment, the predetermined vehicle is obtained in the video image obtained from shooting and is occurred
At least one video segment；
(34) according to the driving information and at least one described video segment, the car of the predetermined vehicle is determined
Image or graphic feature of the trade mark under the reference light degree of illuminating, remember the image or graphic feature to refer to license plate number information.
Further, the driving information includes the geographical location information and correspondingly that each in the path that vehicle passes through is put
At the time of.
Further, the natural lighting brightness is detected by optical sensor and obtained.
Further, the step (31) includes：
(311) positioner and communicator are set for predetermined vehicle；
(312) information of vehicles of the predetermined vehicle is solidified in memory；
(313) the predetermined vehicle recognizes that server sends it and positioned in real time by its communicator to license plate number
Geographical location information and information of vehicles that device is obtained.
Further, the information of vehicles includes the license plate number of vehicle.
Further, the positioner includes GPS device.
Further, the communicator is radio communication device.
Further, the radio communication device recognizes that server leads to by 3G, 4G or GPRS network with the license plate number
Letter.
Further, the step (4) includes：
(41) vehicle for obtaining license plate number to be identified is taken the time or moment of video image, is designated as the moment to be identified；
(42) vehicle for obtaining license plate number to be identified is taken the geographical location information of video image, is designated as position to be identified
Information；
(43) according to the moment to be identified and positional information to be identified, recognize whois lookup to be identified in license plate number
The corresponding position of positional information or region, at the moment to be identified, the predetermined vehicle of appearance；
(44) when finding in the corresponding position of positional information to be identified or region, at the moment to be identified, appearance
During the predetermined vehicle, the reference license plate number information of the predetermined vehicle is determined, this identification is used as
With reference to license plate number information；When to find, according to the natural lighting brightness and the moment to be identified, lookup is treated with described
The predetermined vehicle of the appearance in the corresponding position of identification positional information or the nearest position in region or region is simultaneously determined
It refers to license plate number information, as the reference license plate number information of this identification, and the reference license plate number of this identification is believed
Breath includes the reference license board information of at least one predetermined vehicle；
(45) at least one included according to the reference license plate number information as this identification is described predetermined
The reference license board information of vehicle, recognizes the license plate number of the vehicle of license plate number to be identified successively.
The beneficial effects of the invention are as follows：
(1) present invention can avoid directly going to recognize license plate number using complicated algorithm, so as to reduce identification car plate
Number server cost used.
(2) alleged predetermined vehicle and unspecific vehicle in the present invention, but common vehicle can.Energy
The positioning navigating device and the mobile phone of driver being enough equipped with extensively by Current vehicle, are provided for license plate number identification server
Data sharing, so as to reduce the positioning and the outfit cost of communicator and maintenance of the predetermined vehicle as reference
Cost, and number and threshold that vehicle is provided as reference information are improved, be conducive to effective integration existing resource and honest and clean
Big data in terms of its driving information associated with its license plate number provided as predetermined vehicle is provided valency.
(3) can the reference license plate number information based on predetermined vehicle, be the car plate of the vehicle of license plate number to be identified
Number identification provides direct image graphics object of reference, reduces the complexity of license plate number identification, improves the standard of license plate number identification
True rate.
(4) algorithm for pattern recognition used, can expend the money of image processing server based on object of reference low land as far as possible
Source, so as to significantly reduce the time needed for license plate number is compared based on pixel and machine purchase, maintenance cost.
Brief description of the drawings
Fig. 1 shows the FB(flow block) of the license plate number recognition methods in the traffic video image according to the present invention.
Embodiment
As shown in figure 1, there is provided a kind of identification of the license plate number in traffic video image according to a preferred embodiment of the invention
Method, including：
(1) shoot video image in predetermined shooting area (for example, crossroad) and obtain the shooting area
Geographical location information, the video image includes the traffic video image that pending license plate number is recognized；
(2) natural lighting brightness is recorded while video image is shot；
(3) obtain and refer to license plate number information；
(4) based on the video image for shooting acquisition, by means of the natural lighting brightness and the reference license plate number information,
Recognize the license plate number in the video image.
Preferably, the step (3) includes：
(31) vehicle with predetermined vehicle, the video image that is taken is communicated, and obtains its driving information；
(32) according to the geographical location information of the shooting area, determine the predetermined vehicle by the region
Time or moment, it is designated as referring to the moment, and remembers that with this be reference light degree of illuminating with reference to moment corresponding natural lighting brightness；
(33) according to the reference moment, the predetermined vehicle is obtained in the video image obtained from shooting and is occurred
At least one video segment；
(34) according to the driving information and at least one described video segment, the car of the predetermined vehicle is determined
Image or graphic feature of the trade mark under the reference light degree of illuminating, remember the image or graphic feature to refer to license plate number information.
Preferably, the driving information includes the geographical location information and corresponding that each in the path that vehicle passes through is put
Moment.
Preferably, the natural lighting brightness is detected by optical sensor and obtained.
Preferably, the step (31) includes：
(311) positioner and communicator are set for predetermined vehicle；
(312) information of vehicles of the predetermined vehicle is solidified in memory；
(313) the predetermined vehicle recognizes that server sends it and positioned in real time by its communicator to license plate number
Geographical location information and information of vehicles that device is obtained.
Preferably, the information of vehicles includes the license plate number of vehicle.
Preferably, the positioner includes GPS device.
Preferably, the communicator is radio communication device.
Preferably, the radio communication device recognizes server communication by 3G, 4G or GPRS network with the license plate number.
Preferably, the step (4) includes：
(41) vehicle for obtaining license plate number to be identified is taken the time or moment of video image, is designated as the moment to be identified；
(42) vehicle for obtaining license plate number to be identified is taken the geographical location information of video image, is designated as position to be identified
Information；
(43) according to the moment to be identified and positional information to be identified, recognize whois lookup to be identified in license plate number
The corresponding position of positional information or region, at the moment to be identified, the predetermined vehicle of appearance；
(44) when finding in the corresponding position of positional information to be identified or region, at the moment to be identified, appearance
During the predetermined vehicle, the reference license plate number information of the predetermined vehicle is determined, this identification is used as
With reference to license plate number information；When to find, according to the natural lighting brightness and the moment to be identified, lookup is treated with described
The predetermined vehicle of the appearance in the corresponding position of identification positional information or the nearest position in region or region is simultaneously determined
It refers to license plate number information, as the reference license plate number information of this identification, and the reference license plate number of this identification is believed
Breath includes the reference license board information of at least one predetermined vehicle；
(45) at least one included according to the reference license plate number information as this identification is described predetermined
The reference license board information of vehicle, recognizes the license plate number of the vehicle of license plate number to be identified successively.The step is specifically included：
(451) using the license plate number in reference license plate number information as process object, to reference to the image graph in license plate number information
Shape characteristic information carries out figure segmentation, obtains multiple reference car plate character codes and is referred to those with reference to car plate character code difference is corresponding
Characters on license plate image；
(452) according to traffic video image vehicle, that shoot acquisition of the comparison information to license plate number to be identified, it is somebody's turn to do
The image information of the license plate number region of vehicle to be identified, the image information includes the character figure of multiple license plate numbers to be identified
Picture；
(453) according to this with reference to license plate number information from predetermined vehicle driving trace and car plate to be identified
Number the driving trace of vehicle be compared；
(454) according to the object information of the comparison, it is determined that type and parameter to carrying out image procossing with reference to car plate character code,
The type includes stretching and rotated, and the parameter includes stretching yardstick and the anglec of rotation；
(456) type and parameter handled according to described image, the figure carried out to the multiple reference characters on license plate image
As processing, and compared one by one with the character picture of the license plate number to be identified, until similarity therebetween is higher than pre-
If threshold value, then a word in license plate number to be identified is used as using the corresponding reference car plate character code of reference characters on license plate image now
Code.
It is the purpose to illustrate for the narration that presently preferred embodiments of the present invention is made above, and is not intended to limit essence of the invention
It is really disclosed form, it is possible based on teaching above or to make an amendment or change from embodiments of the invention study
, embodiment is for explanation principle of the invention and allows those skilled in the art to be existed with various embodiments using the present invention
Select and describe in practical application, technological thought of the invention attempts to be determined by claim and its equalization.
Claims (10)
1. the license plate number recognition methods in a kind of traffic video image, it is characterised in that including：
(1) shoot video image in predetermined shooting area and obtain the geographical location information of the shooting area, it is described to regard
Frequency image includes the traffic video image that pending license plate number is recognized；
(2) natural lighting brightness is recorded while video image is shot；
(3) obtain and refer to license plate number information；
(4) based on the video image for shooting acquisition, by means of the natural lighting brightness and the reference license plate number information, identification
License plate number in the video image.
2. according to the method described in claim 1, it is characterised in that the step (3) includes：
(31) vehicle with predetermined vehicle, the video image that is taken is communicated, and obtains its driving information；
(32) according to the geographical location information of the shooting area, determine that the predetermined vehicle passes through the time in the region
Or the moment, it is designated as referring to the moment, and remember that with this be reference light degree of illuminating with reference to moment corresponding natural lighting brightness；
(33) according to the reference moment, from shoot obtain that the predetermined vehicle occurs in the video image that obtains to
A few video segment；
(34) according to the driving information and at least one described video segment, the license plate number of the predetermined vehicle is determined
Image or graphic feature under the reference light degree of illuminating, remember the image or graphic feature to refer to license plate number information.
3. method according to claim 2, it is characterised in that the driving information includes each in the path that vehicle passes through
At the time of the geographical location information and correspondence of point.
4. according to the method described in claim 1, it is characterised in that the natural lighting brightness is obtained by optical sensor detection
.
5. method according to claim 2, it is characterised in that the step (31) includes：
(311) positioner and communicator are set for predetermined vehicle；
(312) information of vehicles of the predetermined vehicle is solidified in memory；
(313) the predetermined vehicle recognizes that server sends its positioner in real time by its communicator to license plate number
The geographical location information and information of vehicles of acquisition.
6. method according to claim 5, it is characterised in that the information of vehicles includes the license plate number of vehicle.
7. method according to claim 5, it is characterised in that the positioner includes GPS device.
8. method according to claim 5, it is characterised in that the communicator is radio communication device.
9. method according to claim 8, it is characterised in that the radio communication device passes through 3G, 4G or GPRS network
Server communication is recognized with the license plate number.
10. method according to claim 8, it is characterised in that the step (4) includes：
(41) vehicle for obtaining license plate number to be identified is taken the time or moment of video image, is designated as the moment to be identified；
(42) vehicle for obtaining license plate number to be identified is taken the geographical location information of video image, is designated as position letter to be identified
Breath；
(43) according to the moment to be identified and positional information to be identified, recognize whois lookup in position to be identified in license plate number
The corresponding position of information or region, at the moment to be identified, the predetermined vehicle of appearance；
(44) when finding in the corresponding position of positional information to be identified or region, at the moment to be identified, appearance it is described
During predetermined vehicle, the reference license plate number information of the predetermined vehicle is determined, the reference of this identification is used as
License plate number information；When to find, according to the natural lighting brightness and the moment to be identified, search with it is described to be identified
The predetermined vehicle of the appearance in the nearest position in the corresponding position of positional information or region or region simultaneously determines that it is joined
License plate number information is examined, the reference license plate number information of this identification, the reference license plate number information bag of this identification is used as
Include the reference license board information of at least one predetermined vehicle；
(45) at least one predetermined vehicle included according to the reference license plate number information as this identification
Reference license board information, the license plate number of the vehicle of license plate number to be identified is recognized successively.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710442152.0A CN107256397A (en) | 2017-06-13 | 2017-06-13 | License plate number recognition methods in traffic video image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710442152.0A CN107256397A (en) | 2017-06-13 | 2017-06-13 | License plate number recognition methods in traffic video image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107256397A true CN107256397A (en) | 2017-10-17 |
Family
ID=60024017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710442152.0A Pending CN107256397A (en) | 2017-06-13 | 2017-06-13 | License plate number recognition methods in traffic video image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107256397A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9137498B1 (en) * | 2011-08-16 | 2015-09-15 | Israel L'Heureux | Detection of mobile computing device use in motor vehicle |
| CN105139659A (en) * | 2015-10-09 | 2015-12-09 | 东方网力科技股份有限公司 | Vehicle license plate recognition method and device |
| CN105321350A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Method and device for detection of fake plate vehicles |
| CN105405298A (en) * | 2015-12-24 | 2016-03-16 | 浙江宇视科技有限公司 | Identification method of license plate identifications and device |
| CN107273848A (en) * | 2017-06-13 | 2017-10-20 | 成都尽知致远科技有限公司 | Video analysis method based on Number Reorganization |
-
2017
- 2017-06-13 CN CN201710442152.0A patent/CN107256397A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9137498B1 (en) * | 2011-08-16 | 2015-09-15 | Israel L'Heureux | Detection of mobile computing device use in motor vehicle |
| CN105321350A (en) * | 2014-08-05 | 2016-02-10 | 北京大学 | Method and device for detection of fake plate vehicles |
| CN105139659A (en) * | 2015-10-09 | 2015-12-09 | 东方网力科技股份有限公司 | Vehicle license plate recognition method and device |
| CN105405298A (en) * | 2015-12-24 | 2016-03-16 | 浙江宇视科技有限公司 | Identification method of license plate identifications and device |
| CN107273848A (en) * | 2017-06-13 | 2017-10-20 | 成都尽知致远科技有限公司 | Video analysis method based on Number Reorganization |
Non-Patent Citations (1)
| Title |
|---|
| 袁宁: "基于 Android 智能手机的车辆远程监控系统的设计与开发", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110660254B (en) | Traffic signal lamp detection and intelligent driving method and device, vehicle and electronic equipment | |
| CN106709436B (en) | Track traffic panoramic monitoring-oriented cross-camera suspicious pedestrian target tracking system | |
| De Charette et al. | Real time visual traffic lights recognition based on spot light detection and adaptive traffic lights templates | |
| Gurghian et al. | Deeplanes: End-to-end lane position estimation using deep neural networksa | |
| CN103824066B (en) | A kind of licence plate recognition method based on video flowing | |
| Zhang et al. | Ripple-GAN: Lane line detection with ripple lane line detection network and Wasserstein GAN | |
| CN107992819B (en) | Method and device for determining vehicle attribute structural features | |
| Romdhane et al. | An improved traffic signs recognition and tracking method for driver assistance system | |
| CN112825192B (en) | Object identification system and method based on machine learning | |
| US20200200545A1 (en) | Method and System for Determining Landmarks in an Environment of a Vehicle | |
| CN112241969A (en) | Target detection tracking method and device based on traffic monitoring video and storage medium | |
| CN111832388B (en) | Method and system for detecting and identifying traffic sign in vehicle running | |
| CN114998815B (en) | Traffic vehicle identification tracking method and system based on video analysis | |
| Liu et al. | Multi-type road marking recognition using adaboost detection and extreme learning machine classification | |
| Cai et al. | Real-time arrow traffic light recognition system for intelligent vehicle | |
| CN110371016A (en) | The distance estimations of front lamp of vehicle | |
| CN107506753B (en) | Multi-vehicle tracking method for dynamic video monitoring | |
| Duan et al. | An end to end recognition for license plates using convolutional neural networks | |
| Joy et al. | Real time road lane detection using computer vision techniques in python | |
| CN108182393A (en) | A kind of automobile and its front truck tracking and system of application | |
| Ghahremannezhad et al. | Automatic road detection in traffic videos | |
| Arthi et al. | Object detection of autonomous vehicles under adverse weather conditions | |
| Moizumi et al. | Traffic light detection considering color saturation using in-vehicle stereo camera | |
| Liu et al. | Real-time traffic light recognition based on smartphone platforms | |
| CN107273848A (en) | Video analysis method based on Number Reorganization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171017 |