CN102867418A - Method and device for judging license plate identification accuracy - Google Patents
Method and device for judging license plate identification accuracy Download PDFInfo
- Publication number
- CN102867418A CN102867418A CN2012103415252A CN201210341525A CN102867418A CN 102867418 A CN102867418 A CN 102867418A CN 2012103415252 A CN2012103415252 A CN 2012103415252A CN 201210341525 A CN201210341525 A CN 201210341525A CN 102867418 A CN102867418 A CN 102867418A
- Authority
- CN
- China
- Prior art keywords
- license plate
- picture
- background
- characters
- automatically
- 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
Links
Images
Abstract
The invention provides a method for judging license plate identification accuracy, which includes the following steps: automatically constructed license plate characters are added in a license plate frame picture, so that a license plate picture is formed; a background picture is automatically acquired from a background picture library, and is a background picture from which a license plate is removed so that a rectangular blank area is formed; after being resized according to a certain proportion, the license plate picture is automatically added in the rectangular blank area of the background picture, so that an image to be identified is synthesized; the synthesized image to be identified is inputted into a testing device, and the testing device carries out license plate identification on the image to be identified according to a license plate identification algorithm and outputs identified license plate characters; and the license plate characters identified by the testing device are automatically compared with the constructed license plate characters in order to determine the accuracy of the license plate identification algorithm. By flexibly constructing license plate characters and mosaicing license plates and a lot of background pictures, the method can construct license plate data to be tested and carry out automated comparison and analysis, thus increasing the controllability and efficiency of testing.
Description
Technical field
The present invention relates to field of video monitoring, relate in particular to the method and apparatus of the accuracy judgement of license plate recognition technology.
Background technology
License plate recognition technology is computer vision and the important subject of mode identification technology in modern intelligent transportation system, is to realize the intelligentized important step of traffic administration.It is take the intelligent identifying system of the technology such as Digital Image Processing, pattern-recognition, computer vision as the basis.It utilizes each automobile that unique number-plate number is arranged, and carries out the identification of the number-plate number by the vehicle image of shot by camera.In the situation that does not affect vehicle condition, computing machine is finished the identification of car plate automatically, thereby can reduce the traffic administration work complexity.Car plate identification is applied in electronic police, the public security bayonet etc. usually, is used for information of vehicles is detected and statistics.
The test of car plate identification accuracy normally needs camera pedestal to be set on the pole of actual road to carry out image capture, then by Recognition Algorithm of License Plate the image of catching is carried out car plate identification, the output recognition result, the last artificial result who one by one Recognition Algorithm of License Plate is exported and the car plate result of eye recognition compare the accuracy rate of statistics Recognition Algorithm of License Plate.
There is obvious shortcoming in above-mentioned method of testing: test environment is built more complicated, need to set up the test environments such as video camera at real crossing; For car plate to be identified, can only rely on the car natural flow to carry out data acquisition, can't clearly test the car plate of specific character combination; Can't automatic decision and statistics, all need rely on human eye to carry out the comparison of picture for every pictures, workload is large, and testing efficiency is low.
Summary of the invention
The invention provides a kind of method of judging car plate identification accuracy, the method comprises the steps: A, reads the license plate frame picture, and the characters on license plate that stack is constructed automatically in the license plate frame picture forms the car plate picture; B, from the Background valut automatic acquisition background picture, this background picture forms the background picture of rectangle white space for rejecting car plate; C, with the car plate picture synthetic image to be identified in the rectangle white space of background picture that automatically is added to behind the convergent-divergent according to a certain percentage; D, the image to be identified that will synthesize are input to equipment to be tested, treat recognition image carries out car plate identification output identification according to Recognition Algorithm of License Plate characters on license plate by this equipment to be tested; E, the characters on license plate that testing apparatus is identified and the characters on license plate of structure compare automatically, to determine the precision of Recognition Algorithm of License Plate.
Preferably, this scaling is pic_zoom, pic_zoom=(L1*W1)/(L2*W2), and wherein L1 and W1 are respectively the length of background picture rectangle and wide, and L2 and W2 are respectively the length of car plate picture and wide.
Preferably, this car plate picture keeps behind the convergent-divergent in the rectangle white space of the background picture that automatically is added to consistent with the background picture pixel according to a certain percentage.
Preferably, the background picture in the Background valut is the picture of actual photographed on the road.
The present invention also provides a kind of device of judging car plate identification accuracy, it is characterized in that this device comprises:
Image synthesis unit is used for reading the license plate frame picture, and the characters on license plate that stack is constructed automatically in the license plate frame picture forms the car plate picture; And automatic acquisition background picture from the Background valut, this background picture form the background picture of rectangle white space for rejecting car plate; And with the car plate picture synthetic image to be identified in the rectangle white space of background picture that automatically is added to behind the convergent-divergent according to a certain percentage; Image output module, the image to be identified that is used for synthesizing is input to equipment to be tested, treats recognition image carries out car plate identification output identification according to Recognition Algorithm of License Plate characters on license plate by this equipment to be tested; Discrimination module compares automatically for the characters on license plate that testing apparatus is identified and the characters on license plate of structure, to determine the precision of Recognition Algorithm of License Plate.
Beneficial effect of the present invention: by the structure of characters on license plate flexibly, the splicing of car plate and numerous background pictures consists of car plate data to be tested, and carries out robotization relatively and analyze, and has improved controllability and the efficient of test.
Description of drawings
Fig. 1 is a kind of process flow diagram of automatically the Recognition Algorithm of License Plate accuracy being tested of the present invention.
Fig. 2 is a kind of license plate frame picture.
Fig. 3 is a kind of car plate picture.
Fig. 4 is apparatus of the present invention module map.
Embodiment
Below in conjunction with Fig. 1 the specific embodiment of the invention is described.Fig. 1 is a kind of process flow diagram of automatically the Recognition Algorithm of License Plate accuracy being tested.
Step 11, structure car plate picture.
The car plate picture comprises two parts content: license plate frame picture (ginseng Fig. 2) and characters on license plate.Pre-save license plate frame picture.The license plate frame picture comprise blue license plate frame (such as the civil small-scale automobile), yellow license plate frame (as in, large car and motorcycle), black license plate frame (such as foreign nationality's car) etc.Characters on license plate is constructed automatically, and the rule of structure characters on license plate is: 1 Chinese character (provinces, autonomous regions and municipalities' abbreviation etc.)+1 English alphabet (the machine-operated code name of dealing out the cards)+blank character ()+5 numeral and English alphabet mix the character that forms.The license plate frame picture of reading and saving is superimposed upon formation car plate picture on the license plate frame picture with the characters on license plate of automatically constructing.Car plate picture please refer to the drawing 3 of a formation.
Step 12, background extraction picture
Automatic acquisition background picture from the Background valut, this background picture be on the road actual photographed but reject the background picture of license plate area.For the test effect that obtains, the background picture of various actual conditions can as much as possiblely be set.Cross the picture of car such as the picture of crossing car daytime, evening; The picture of cart, the picture of dolly; The fine picture of crossing car, the rainy day crosses picture of car etc.The license plate area of vehicle is disallowable in these background pictures falls and forms the white space of rectangle.
Step 13, synthetic image to be identified
Automatically the rectangle white space that is superimposed upon background picture behind the car plate picture bi-directional scaling is formed image to be identified.The space rectangles area size of car plate may be inconsistent in car plate picture and the background picture, so at the scaling that carries out need to calculating when image synthesizes the car plate picture.Concrete computing method are: the long L1 of background extraction picture rectangle and wide W1, obtain long L2 and the wide W2 of car plate picture; Calculate scaling pic_zoom=(L1*W1)/(L2*W2).The car plate picture is carried out convergent-divergent according to scaling pic_zoom, and maintenance is consistent with the background picture pixel.The car plate picture that convergent-divergent is the complete background picture that is added to forms final image to be identified.
Step 14, car plate identification
Synthetic image to be identified is input to equipment to be tested, is carried out the characters on license plate of car plate identification output identification by this equipment synthetic images to be tested according to Recognition Algorithm of License Plate.
Step 15, accuracy are judged
The characters on license plate that testing apparatus is identified and the characters on license plate of structure compare automatically, judge whether car plate identification is accurate.
Step 11～step 15 is a test result, repeated execution of steps 11～step 15, all construct each time different characters on license plate, choose different background pictures, all sidedly the various car plates under the diversity of settings are identified, the car plate with recognition result and structure relatively obtains some test results again.Test result is added up automatically, and after carrying out 1000 tests, correct number of times is 950, and that equipment to be tested is 95% to the accuracy rate of car plate identification.
Because characters on license plate for automatically to construct according to rule, so can be used for various functions and the abnormality test of characters on license plate input combination, need not to wait for that actual license plate occurs.The car plate of structure and the combination of diversity of settings scene can form a large amount of resolution charts easily, make things convenient for function and scrnario testing.Whole test process, the input of test data is by procedure auto-control, and the output of test data is compared automatically by program, has improved testing efficiency.
Based on same design, the present invention also provides a kind of device of judging car plate identification accuracy.Such as Fig. 4, this device comprises image synthesis unit, image output module and discrimination module.
Image synthesis unit is used for reading the license plate frame picture, and the characters on license plate that stack is constructed automatically in the license plate frame picture forms the car plate picture; And automatic acquisition background picture from the Background valut, this background picture form the background picture of rectangle white space for rejecting car plate; And with the car plate picture synthetic image to be identified in the rectangle white space of background picture that automatically is added to behind the convergent-divergent according to a certain percentage.
Image output module, the image to be identified that is used for synthesizing is input to equipment to be tested, treats recognition image carries out car plate identification output identification according to Recognition Algorithm of License Plate characters on license plate by this equipment to be tested.
Discrimination module compares automatically for the characters on license plate that testing apparatus is identified and the characters on license plate of structure, to determine the precision of Recognition Algorithm of License Plate.
The above only is preferred embodiment of the present invention, and is in order to limit the present invention, within the spirit and principles in the present invention not all, any modification of making, is equal to replacement, improvement etc., all should be included within the scope of protection of the invention.
Claims (8)
1. a method of judging car plate identification accuracy is characterized in that the method comprises the steps:
A, read the license plate frame picture, in the license plate frame picture stack automatically the characters on license plate of structure form the car plate picture;
B, from the Background valut automatic acquisition background picture, this background picture forms the background picture of rectangle white space for rejecting car plate;
C, with the car plate picture synthetic image to be identified in the rectangle white space of background picture that automatically is added to behind the convergent-divergent according to a certain percentage;
D, the image to be identified that will synthesize are input to equipment to be tested, treat recognition image carries out car plate identification output identification according to Recognition Algorithm of License Plate characters on license plate by this equipment to be tested;
E, the characters on license plate that testing apparatus is identified and the characters on license plate of structure compare automatically, to determine the precision of Recognition Algorithm of License Plate.
2. the method for claim 1, it is characterized in that described scaling is pic_zoom, pic_zoom=(L1*W1)/(L2*W2), wherein L1 and W1 are respectively the length of background picture rectangle and wide, and L2 and W2 are respectively the length of car plate picture and wide.
3. the method for claim 1 is characterized in that, described car plate picture keeps behind the convergent-divergent in the rectangle white space of the background picture that automatically is added to consistent with the background picture pixel according to a certain percentage.
4. the method for claim 1 is characterized in that, the background picture in the described Background valut is the picture of actual photographed on the road.
5. device of judging car plate identification accuracy is characterized in that this device comprises:
Image synthesis unit is used for reading the license plate frame picture, and the characters on license plate that stack is constructed automatically in the license plate frame picture forms the car plate picture; And automatic acquisition background picture from the Background valut, this background picture form the background picture of rectangle white space for rejecting car plate; And with the car plate picture synthetic image to be identified in the rectangle white space of background picture that automatically is added to behind the convergent-divergent according to a certain percentage;
Image output module, the image to be identified that is used for synthesizing is input to equipment to be tested, treats recognition image carries out car plate identification output identification according to Recognition Algorithm of License Plate characters on license plate by this equipment to be tested;
Discrimination module compares automatically for the characters on license plate that testing apparatus is identified and the characters on license plate of structure, to determine the precision of Recognition Algorithm of License Plate.
6. method as claimed in claim 5, it is characterized in that described scaling is pic_zoom, pic_zoom=(L1*W1)/(L2*W2), wherein L1 and W1 are respectively the length of background picture rectangle and wide, and L2 and W2 are respectively the length of car plate picture and wide.
7. method as claimed in claim 5 is characterized in that, described car plate picture keeps behind the convergent-divergent in the rectangle white space of the background picture that automatically is added to consistent with the background picture pixel according to a certain percentage.
8. method as claimed in claim 5 is characterized in that, the background picture in the described Background valut is the picture of actual photographed on the road.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210341525.2A CN102867418B (en) | 2012-09-14 | 2012-09-14 | Method and device for judging license plate identification accuracy |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201210341525.2A CN102867418B (en) | 2012-09-14 | 2012-09-14 | Method and device for judging license plate identification accuracy |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102867418A true CN102867418A (en) | 2013-01-09 |
| CN102867418B CN102867418B (en) | 2014-10-22 |
Family
ID=47446270
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201210341525.2A Active CN102867418B (en) | 2012-09-14 | 2012-09-14 | Method and device for judging license plate identification accuracy |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102867418B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104424624A (en) * | 2013-08-28 | 2015-03-18 | 中兴通讯股份有限公司 | Image synthesis optimization method and device |
| CN105046978A (en) * | 2015-08-27 | 2015-11-11 | 宝鸡市交通信息工程研究所 | Vehicle information LED display system possessing realistic effect and realization method |
| CN105096354A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN106462770A (en) * | 2014-04-30 | 2017-02-22 | 莱克斯真株式会社 | Device for converting the vehicle number into the image, method for controlling the same and vehicle number reading system using the same |
| CN106779089A (en) * | 2016-12-13 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method of testing and device |
| CN108229466A (en) * | 2016-12-15 | 2018-06-29 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method and device |
| CN108415915A (en) * | 2017-02-10 | 2018-08-17 | 深圳怡化电脑股份有限公司 | A kind of proof of algorithm method and device based on bank note data |
| CN108875837A (en) * | 2018-06-27 | 2018-11-23 | 北京旷视科技有限公司 | Device and method for test target identification equipment |
| CN108932853A (en) * | 2018-06-22 | 2018-12-04 | 安徽科力信息产业有限责任公司 | A kind of method and device recording more motor vehicle illegal parking behaviors |
| CN110647705A (en) * | 2019-08-14 | 2020-01-03 | 平安普惠企业管理有限公司 | Picture synthesis method, device, equipment and storage medium |
| CN110659632A (en) * | 2019-09-29 | 2020-01-07 | 公安部交通管理科学研究所 | System and method for testing motor vehicle number plate identification performance of traffic technology monitoring equipment based on image block assignment |
| CN111586319A (en) * | 2020-05-27 | 2020-08-25 | 北京百度网讯科技有限公司 | Video processing method and device |
| CN112101349A (en) * | 2020-09-01 | 2020-12-18 | 北京智芯原动科技有限公司 | License plate sample generation method and device |
| CN112560551A (en) * | 2019-09-25 | 2021-03-26 | 西门子（中国）有限公司 | System, method and device for license plate recognition |
| CN113096111A (en) * | 2021-04-25 | 2021-07-09 | 深圳市永丰盈电子有限公司 | Material detection method, system, computer program product and readable storage medium |
| CN113345037A (en) * | 2021-06-07 | 2021-09-03 | 重庆紫光华山智安科技有限公司 | Automatic testing method, system, medium and terminal for motor vehicle algorithm indexes |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5381155A (en) * | 1993-12-08 | 1995-01-10 | Gerber; Eliot S. | Vehicle speeding detection and identification |
| US5608391A (en) * | 1995-05-11 | 1997-03-04 | Minnesota Mining And Manufacturing Company | Electronic license plate architecture |
| US5657008A (en) * | 1995-05-11 | 1997-08-12 | Minnesota Mining And Manufacturing Company | Electronic license plate having a secure identification device |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN101673338A (en) * | 2009-10-09 | 2010-03-17 | 南京树声科技有限公司 | Fuzzy license plate identification method based on multi-angle projection |
| CN101872416A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Vehicle license plate recognition method and system of road image |
-
2012
- 2012-09-14 CN CN201210341525.2A patent/CN102867418B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5381155A (en) * | 1993-12-08 | 1995-01-10 | Gerber; Eliot S. | Vehicle speeding detection and identification |
| US5608391A (en) * | 1995-05-11 | 1997-03-04 | Minnesota Mining And Manufacturing Company | Electronic license plate architecture |
| US5657008A (en) * | 1995-05-11 | 1997-08-12 | Minnesota Mining And Manufacturing Company | Electronic license plate having a secure identification device |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| CN101673338A (en) * | 2009-10-09 | 2010-03-17 | 南京树声科技有限公司 | Fuzzy license plate identification method based on multi-angle projection |
| CN101872416A (en) * | 2010-05-06 | 2010-10-27 | 复旦大学 | Vehicle license plate recognition method and system of road image |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104424624A (en) * | 2013-08-28 | 2015-03-18 | 中兴通讯股份有限公司 | Image synthesis optimization method and device |
| CN104424624B (en) * | 2013-08-28 | 2018-04-10 | 中兴通讯股份有限公司 | A kind of optimization method and device of image synthesis |
| CN106462770A (en) * | 2014-04-30 | 2017-02-22 | 莱克斯真株式会社 | Device for converting the vehicle number into the image, method for controlling the same and vehicle number reading system using the same |
| CN105096354A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN105046978A (en) * | 2015-08-27 | 2015-11-11 | 宝鸡市交通信息工程研究所 | Vehicle information LED display system possessing realistic effect and realization method |
| CN105046978B (en) * | 2015-08-27 | 2017-10-31 | 宝鸡市交通信息工程研究所 | A kind of information of vehicles LED display system and implementation method with vivid effect |
| CN106779089A (en) * | 2016-12-13 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method of testing and device |
| CN108229466A (en) * | 2016-12-15 | 2018-06-29 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method and device |
| CN108229466B (en) * | 2016-12-15 | 2020-12-11 | 杭州海康威视数字技术股份有限公司 | License plate recognition method and device |
| CN108415915A (en) * | 2017-02-10 | 2018-08-17 | 深圳怡化电脑股份有限公司 | A kind of proof of algorithm method and device based on bank note data |
| CN108932853A (en) * | 2018-06-22 | 2018-12-04 | 安徽科力信息产业有限责任公司 | A kind of method and device recording more motor vehicle illegal parking behaviors |
| CN108875837A (en) * | 2018-06-27 | 2018-11-23 | 北京旷视科技有限公司 | Device and method for test target identification equipment |
| CN110647705A (en) * | 2019-08-14 | 2020-01-03 | 平安普惠企业管理有限公司 | Picture synthesis method, device, equipment and storage medium |
| CN112560551A (en) * | 2019-09-25 | 2021-03-26 | 西门子（中国）有限公司 | System, method and device for license plate recognition |
| CN110659632A (en) * | 2019-09-29 | 2020-01-07 | 公安部交通管理科学研究所 | System and method for testing motor vehicle number plate identification performance of traffic technology monitoring equipment based on image block assignment |
| CN111586319A (en) * | 2020-05-27 | 2020-08-25 | 北京百度网讯科技有限公司 | Video processing method and device |
| US11700417B2 (en) | 2020-05-27 | 2023-07-11 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Method and apparatus for processing video |
| CN111586319B (en) * | 2020-05-27 | 2024-04-09 | 北京百度网讯科技有限公司 | Video processing method and device |
| CN112101349A (en) * | 2020-09-01 | 2020-12-18 | 北京智芯原动科技有限公司 | License plate sample generation method and device |
| CN113096111A (en) * | 2021-04-25 | 2021-07-09 | 深圳市永丰盈电子有限公司 | Material detection method, system, computer program product and readable storage medium |
| CN113345037A (en) * | 2021-06-07 | 2021-09-03 | 重庆紫光华山智安科技有限公司 | Automatic testing method, system, medium and terminal for motor vehicle algorithm indexes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102867418B (en) | 2014-10-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102867418B (en) | Method and device for judging license plate identification accuracy | |
| Zhang et al. | CCTSDB 2021: a more comprehensive traffic sign detection benchmark | |
| Omachi et al. | Detection of traffic light using structural information | |
| TWI308725B (en) | ||
| Sulaiman et al. | Development of automatic vehicle plate detection system | |
| Saha et al. | License Plate localization from vehicle images: An edge based multi-stage approach | |
| CN106971185B (en) | License plate positioning method and device based on full convolution network | |
| CN110348463B (en) | Method and device for identifying vehicle | |
| CN103996041A (en) | Vehicle color identification method and system based on matching | |
| CN109033175A (en) | A kind of method and system to scheme to search vehicle | |
| CN103530638A (en) | Method for matching pedestrians under multiple cameras | |
| CN113808098A (en) | Road disease identification method and device, electronic equipment and readable storage medium | |
| AG | Development of portable automatic number plate recognition (ANPR) system on Raspberry Pi | |
| CN110889388A (en) | Violation identification method, device, equipment and storage medium | |
| Pramkeaw et al. | Analysis of detecting and interpreting warning signs for distance of cars using analyzing the license plate | |
| CN111881984A (en) | Target detection method and device based on deep learning | |
| Lin et al. | A lightweight, high-performance multi-angle license plate recognition model | |
| Jain et al. | A hybrid approach for detection and recognition of traffic text sign using MSER and OCR | |
| Jensen et al. | Presenting the multi-view traffic intersection dataset (MTID): A detailed traffic-surveillance dataset | |
| CN111339834B (en) | Method for identifying vehicle driving direction, computer device and storage medium | |
| CN115601717B (en) | Deep learning-based traffic offence behavior classification detection method and SoC chip | |
| CN106462770A (en) | Device for converting the vehicle number into the image, method for controlling the same and vehicle number reading system using the same | |
| Nguyen et al. | Triple detector based on feature pyramid network for license plate detection and recognition system in unusual conditions | |
| CN114170565A (en) | Image comparison method and device based on unmanned aerial vehicle aerial photography and terminal equipment | |
| CN106920398A (en) | A kind of intelligent vehicle license plate recognition system |
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 |