CN106778743A - A kind of improved car plate voting method and device - Google Patents
A kind of improved car plate voting method and device Download PDFInfo
- Publication number
- CN106778743A CN106778743A CN201611141388.2A CN201611141388A CN106778743A CN 106778743 A CN106778743 A CN 106778743A CN 201611141388 A CN201611141388 A CN 201611141388A CN 106778743 A CN106778743 A CN 106778743A
- Authority
- CN
- China
- Prior art keywords
- car plate
- result
- level
- results
- grades
- 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
- G06V20/60—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)
- Character Discrimination (AREA)
Abstract
The invention discloses a kind of improved car plate voting method and device, the method includes:Obtain multiple license plate recognition results;The multiple license plate recognition result is carried out into grade distinguishing, is divided into one-level car plate result, two grades of car plate results and three-level car plate result, and the multiple license plate recognition result is added in queue;One-level car plate result is selected from queue, if the number of one-level car plate result is more than or equal to preset value, car plate ballot is carried out in all of one-level car plate result, obtain car plate voting results.The method realizes improving the accuracy of voting results.
Description
Technical field
The present invention relates to field of license plate recognition, more particularly to a kind of improved car plate voting method and device.
Background technology
At present, Car license recognition equipment on the market mainly has single frames to recognize and two kinds of video flowing identification.Single frames identification is main
Combined with ground sense, trigger video capture image after the ground sense of vehicle pressure and recognize the number-plate number of present frame.Video flowing identification is logical
Cross and continuously acquire image and recognize, then the comprehensive single frames result for obtaining obtains last answer.The result of video flowing identification is general
Can be stored in a queue, be as a result managed generally by Voting Algorithm.Traditional car plate voting method includes three
Part:Queue management, vote condition judgment and voting process.
Queue management is:Per road one queue of maintenance of camera, the image of each frame input all carries out Car license recognition, if
The result of Car license recognition is reliably to be added in queue.License plate recognition result queue uses first in first out strategy, no matter team
Whether row have voted, as long as there is the recognition result will to be added to queue.After queue full, then there is the new result to come in just most
An early record is emptied, and preserves current result.It should be noted that only working as new car or not knowing for a period of time
Other result can just empty queue.The length of queue can be set by user oneself, for example, the length of queue is set to 10, i.e., only
Voted using 10 effective recognition results.It is accurate if necessary to recognize, the length of queue can be set to 20, even if
Voted with more results.Queue is long to mean that the more results of needs, and speed will be slower than short queue.
Ballot condition judgment be:The each frame of system can all check whether the condition of ballot meets, and the condition of ballot is included newly
Car comes, and vehicle disappears, and frame number is enough and signal is felt on ground.Queue will be triggered when one in above-mentioned condition meets
Voted.Feel signal with will being produced when vehicle feels upper with being parked in, system preserves present frame, captured as ground sense and tied
Really.The method voted using video flowing, before sense signal in ground comes, if the result for having had video flowing to vote, is ignored as ground
Sense signal, while feel the image of candid photograph with abandoning.When signal is felt with receiving, if not having the knot that video flowing is voted also
Really, certain frame number (feel time delay ballot frame number) can be waited, in this process, if the result of video flowing ballot, is ignored
Ground sense signal, while feel the image of candid photograph with abandoning.Wait frame number much of that, still there is no video flowing voting results, trigger video
Stream ballot.
Voting process is:For meeting ballot condition and also unpolled queue, ballot is performed.If voted successfully,
Labeled as having voted.If present frame ballot failure, next frame is just waited to continue to throw, if until car disappears or new car arrives
Still without voting results, the car plate just leaks.For meeting ballot condition and labeled as the queue voted, ignoring the throwing
Ticket condition.
When carrying out car plate ballot at present, the weight of all of result is the same, i.e., all of result is thrown together
Ticket.In car plate voting process, either confidence level it is high or the low license plate recognition result of confidence level, can be placed on queue
In voted, wait queue full, to queue in all license plate recognition results vote.In general, closer to shooting
The result of machine, car plate is bigger and more clear, and the confidence level of character can be higher in theory.For the preferable scene of live illumination,
Even if car plate is distant, but the confidence level of character is also very high.If work as running into for example live illumination of some environmental factors very
During difference, for the scene that live light is photoed badly, when a long way off, for Chinese character, discrimination can be slightly worse, close
When, Chinese character can be more reliable, and because photo badly license plate recognition result that confidence level in queue can be caused low of light becomes many, it is so right
Recognition result in one queue is voted, then final vote result is the probability meeting of the low license plate recognition result of confidence level
Increase, the voting results of mistake most probably occur, the accuracy of voting results is relatively low.
The content of the invention
It is an object of the invention to provide a kind of improved car plate voting method and device, to realize improving the standard of voting results
True property.
In order to solve the above technical problems, the present invention provides a kind of improved car plate voting method, the method includes:
Obtain multiple license plate recognition results;
The multiple license plate recognition result is carried out into grade distinguishing, is divided into one-level car plate result, two grades of car plate results and three
Level car plate result, and the multiple license plate recognition result is added in queue;
One-level car plate result is selected from queue, if the number of one-level car plate result is more than or equal to preset value, all
One-level car plate result in carry out car plate ballot, obtain car plate voting results.
Preferably, methods described also includes:
If the number of one-level car plate result is less than preset value, two grades of car plate results are selected from queue;If one-level car plate
The number of result and two grades of summations of the number of car plate result are more than or equal to preset value, in all of one-level car plate result and two grades
Car plate ballot is carried out in car plate result, car plate voting results are obtained.
Preferably, methods described also includes:
If the number of one-level car plate result and two grades of summations of the number of car plate result are less than preset value, the institute in queue
Car plate ballot is carried out in some license plate recognition results, car plate voting results are obtained.
Preferably, the confidence level of one-level car plate result more than the confidence levels of two grades of car plate results, put by two grades of car plate results
Confidence level of the reliability more than three-level car plate result.
Preferably, methods described also includes:
Judge whether the Chinese character in car plate voting results meets identification and require, if it is not, by the Chinese character in car plate voting results
It is substituted for default character.
The present invention also provides a kind of improved car plate balloting device, and the device includes:
Acquisition module, for obtaining multiple license plate recognition results;
Discriminating module, for the multiple license plate recognition result to be carried out into grade distinguishing, be divided into one-level car plate result, two grades
Car plate result and three-level car plate result, and the multiple license plate recognition result is added in queue;
First vote module, for selecting one-level car plate result from queue, if the number of one-level car plate result is more than
Equal to preset value, car plate ballot is carried out in all of one-level car plate result, obtain car plate voting results.
Preferably, described device also includes:
Second vote module, if being less than preset value for the number of one-level car plate result, selects two grades of cars from queue
Board result;If the number of one-level car plate result and two grades of summations of the number of car plate result are more than or equal to preset value, all of
Car plate ballot is carried out in one-level car plate result and two grades of car plate results, car plate voting results are obtained.
Preferably, described device also includes:
3rd vote module, if the number and two grades of summations of the number of car plate result for one-level car plate result are less than pre-
If value, car plate ballot is carried out in all of license plate recognition result in queue, car plate voting results are obtained.
Preferably, the confidence level of one-level car plate result more than the confidence levels of two grades of car plate results, put by two grades of car plate results
Confidence level of the reliability more than three-level car plate result.
Preferably, described device also includes:
Replacement module, whether the Chinese character for judging in car plate voting results meets identification requires, if it is not, car plate is voted
Chinese character in result is substituted for default character.
A kind of improved car plate voting method provided by the present invention and device, obtain multiple license plate recognition results;By institute
Stating multiple license plate recognition results carries out grade distinguishing, is divided into one-level car plate result, two grades of car plate results and three-level car plate result, and
The multiple license plate recognition result is added in queue;One-level car plate result is selected from queue, if one-level car plate result
Number be more than or equal to preset value, carry out car plate ballot in all of one-level car plate result, obtain car plate voting results.Can
See, license plate recognition result is first divided into Three Estate, and one-level car plate result is confidence level highest in Three Estate, car plate ballot
When first choose one-level car plate result, if one-level car plate result number enough if complete ballot, just use Car license recognition
All one-level car plate results in result are voted, and are voted only with one-level car plate result, and such light is photoed badly can lead
The low license plate recognition result of confidence level is caused in queue all without being chosen as object of voting, then final vote result is that confidence level is high
The probability of license plate recognition result can increase, effectively improve the accuracy of voting results.
Brief description of the drawings
In order to illustrate more clearly about the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing
The accompanying drawing to be used needed for having technology description is briefly described, it should be apparent that, drawings in the following description are only this
Inventive embodiment, for those of ordinary skill in the art, on the premise of not paying creative work, can also basis
The accompanying drawing of offer obtains other accompanying drawings.
Fig. 1 is a kind of flow chart of improved car plate voting method provided by the present invention;
Fig. 2 is improved car plate voting method outline flowchart;
Fig. 3 is a kind of structural representation of improved car plate balloting device provided by the present invention.
Specific embodiment
Core of the invention is to provide a kind of improved car plate voting method and device, to realize improving the standard of voting results
True property.
In order that those skilled in the art more fully understand the present invention program, below in conjunction with the embodiment of the present invention
Accompanying drawing, is clearly and completely described to the technical scheme in the embodiment of the present invention, it is clear that described embodiment is only
A part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art
The every other embodiment obtained under the premise of creative work is not made, belongs to the scope of protection of the invention.
Fig. 1 is refer to, Fig. 1 is a kind of flow chart of improved car plate voting method provided by the present invention, the method bag
Include:
S11:Obtain multiple license plate recognition results;
S12:Multiple license plate recognition results are carried out into grade distinguishing, is divided into one-level car plate result, two grades of car plate results and three
Level car plate result, and multiple license plate recognition results are added in queue;
S13:One-level car plate result is selected from queue, if the number of one-level car plate result is more than or equal to preset value,
Car plate ballot is carried out in all of one-level car plate result, car plate voting results are obtained.
It can be seen that, license plate recognition result is first divided into Three Estate by the method, and one-level car plate result is confidence in Three Estate
Degree highest, car plate vote when first choose one-level car plate result, if one-level car plate result number enough just complete throwing
Ticket, with regard to being voted using all one-level car plate results in license plate recognition result, is voted only with one-level car plate result,
Such light photos badly license plate recognition result that confidence level in queue can be caused low all without being chosen as object of voting, then final to throw
Ticket result is that the probability of confidence level license plate recognition result high can increase, and effectively improves the accuracy of voting results.
Based on the above method, wherein, the confidence level of one-level car plate result is more than two grades of confidence levels of car plate result, two grades of cars
Confidence level of the confidence level of board result more than three-level car plate result.The multiple license plate recognition results obtained in step S11 have been
The license plate recognition result of effect.License plate recognition result is exactly the two field picture for obtained after Car license recognition.
Further, methods described also includes:If the number of one-level car plate result is less than preset value, selected from queue
Two grades of car plate results;If the number of one-level car plate result and two grades of summations of the number of car plate result are more than or equal to preset value,
Car plate ballot is carried out in all of one-level car plate result and two grades of car plate results, car plate voting results are obtained.
Wherein, license plate recognition result segmentation is obtained into 7 characters, each single character is identified again then, obtained
To 7 confidence levels of character.If the confidence level of 7 characters is both greater than 0.9, then the license plate recognition result is rated as one-level
Car plate result.If having 6 confidence levels of character more than 0.9 in 7 characters and the confidence level of other 1 character being more than 0.7,
So the car plate is rated as two grades of car plates.The car plate of other situations is considered as three-level car plate.
Further, methods described also includes:If the number of one-level car plate result and two grades of numbers of car plate result is total
With less than preset value, car plate ballot is carried out in all of license plate recognition result in queue, obtain car plate voting results.
Further, methods described also includes:Judge whether the Chinese character in car plate voting results meets identification and require, if
It is no, the Chinese character in car plate voting results is substituted for default character.
In this way, the method first chooses first-class result, if poll foot first effective car plate point Three Estate when ballot
It is enough just to complete to vote.If poll not enough chooses first-class and second-class result again, if poll enough if complete ballot.If poll
Still it is insufficient to just be voted using whole results.Finally Chinese character result is judged whether by poll again need to use acquiescence
Character is replaced.The method can effectively solve ballot caused by scene causes recognition result confidence level not high due to illumination reason
The problem of error, and improve the discrimination of Chinese character.
Wherein, the scene photoed badly for light, is only voted with the result of the partial frame near video camera, and effect ratio is used
Whole result ballots is more preferable.For the preferable scene of site environment, voted using more results.
Wherein, for most scene, the car plate at scene can, from same province, equipment be pacified more than 90%
Local province province by default is filled, if the confidence level of Chinese Character Recognition is not high, acquiescence province is directly used as identification
As a result, 90% Chinese Character Recognition rate can so be had more than.
With reference to Fig. 2, Fig. 2 is improved car plate voting method outline flowchart.The specific reality of improved car plate voting method
Apply process as follows:
1st, after every frame Video processing, whether effectively the recognition result of present frame is judged, if result effectively, is known car plate
Other result classification, is then added in queue, otherwise terminates present frame operation.
Wherein, the process of the Car license recognition of each frame video includes the positioning to the region comprising car plate, to comprising car plate
The segmentation in region obtains 7 characters, and each single character is identified again then.
Wherein, using the character classifier trained based on SVMs, the character that each segmentation is finished can use word
Symbol grader is identified.The recognition result and the confidence level of the character of the result meeting output character of identification.Confidence level can be with
It is interpreted as grader and judges that current character is the reliability of this class character, confidence level is typically the numerical value between a 0-1, number
Value is more high more reliable.Car plate judges whether reliability by the SVM confidence levels of 7 characters of recognition result.Surveyed by many experiments
Examination, have selected two threshold values, be respectively ThighWith Tlow.Once the confidence level of recognition result has exceeded ThighMean that the result is non-
It is often reliable.The result of identification is in ThighWith TlowBetween, represent that the result has certain probability to malfunction.Recognition result is less than TlowWith regard to table
Show that result is not very reliable.Preferably, the value of Thigh is 0.9, TlowValue be 0.7.
Wherein, the identification of each frame can all count 7 confidence levels of character, if the confidence level of 7 characters has more than 2
Less than Tlow, then the license plate recognition result is considered underproof, will be killed without being saved in queue.Have passed through first
License plate recognition result after screening, if qualified be added in car plate queue.If the confidence level of 7 characters is both greater than
Thigh, then the car plate is rated as one-level car plate.If there is 6 confidence levels of character to be more than T in 7 charactershighAnd in addition
1 confidence level of character is more than Tlow, then the car plate is rated as two grades of car plates.The car plate of other situations is considered as three-level
Car plate.
2nd, judge whether present frame meets ballot condition, if it is satisfied, then carrying out a point three-level ballot.
Wherein, when car plate is voted, one-level car plate is elected first, the number of one-level car plate is counted, as long as one-level
The number of car plate has exceeded the 20% of queue length, it is possible to one-level car plate is used alone and is voted, if voting successfully
Using voting results as queue result.If the number of one-level car plate does not reach 20% or has reached 20%, but votes
Failure, is just used together one-level car plate with two grades of car plates.If one-level car plate adds up long more than queue with two grades of car plates
The 50% of degree, it is possible to voted with two grades of car plates using one-level car plate, using voting results as team if voting successfully
The result of row.If one-level car plate and two grades of car plates add up no more than queue length 50% or reached 50%, but
It is ballot failure, just all car plates is used together and is voted.
Under most of scene, as long as illumination condition is good, video camera install have no problem, queue can be easy for it is full,
And the car plate in queue more than half all can be one-level car plate.Only when live light is photoed badly, video camera is installed not just, or
Occur in that seriously with car, situations such as blocking, just occur that the effective car plate in queue is few, or one-level is few with two grades of car plates
Situation.In this case, select limited good car plate to be voted, often have than carrying out ballot using whole results
Discrimination higher.For example there occurs and block only have 5-10 frame valid frames in possible queue, then selected in limited 5-10 frames
2-3 frame one-level car plates are selected to be voted, can be better than the effect voted using whole results.
3rd, judge whether the Chinese character of current ballot result meets demand, default character is carried out if being unsatisfactory for and replaces defeated again
Go out voting results, otherwise export voting results.
Wherein, after there are the voting results of Chinese character, compare whether poll highest Chinese character gives tacit consent to province, if acquiescence is saved
Part, it is not necessary to it is adjusted again.There are following three kinds of situations to be just replaced:
(1) if the Chinese character in poll highest result is not acquiescence province, but the Chinese in the result high of poll second
Word is acquiescence province, and the second poll high exceedes the 70% of highest poll, just carries out acquiescence province and replaces.
(2) average confidence of the chinese character of statistics poll highest result, if average confidence does not reach Thigh,
So just it is replaced.
(3) if the poll of poll highest result is less than the half of queue length, and poll highest poll is subtracted
20% of the difference of the poll of poll second less than queue length, then be just replaced.
Fig. 3 is refer to, Fig. 3 is a kind of structural representation of improved car plate balloting device provided by the present invention, the device
Including:
Acquisition module 101, for obtaining multiple license plate recognition results;
Discriminating module 102, for multiple license plate recognition results to be carried out into grade distinguishing, be divided into one-level car plate result, two grades
Car plate result and three-level car plate result, and multiple license plate recognition results are added in queue;
First vote module 103, for selecting one-level car plate result from queue, if the number of one-level car plate result is big
In equal to preset value, car plate ballot is carried out in all of one-level car plate result, obtain car plate voting results.
It can be seen that, license plate recognition result is first divided into Three Estate by the device, and one-level car plate result is confidence in Three Estate
Degree highest, car plate vote when first choose one-level car plate result, if one-level car plate result number enough just complete throwing
Ticket, with regard to being voted using all one-level car plate results in license plate recognition result, is voted only with one-level car plate result,
Such light photos badly license plate recognition result that confidence level in queue can be caused low all without being chosen as object of voting, then final to throw
Ticket result is that the probability of confidence level license plate recognition result high can increase, and effectively improves the accuracy of voting results.
Based on said apparatus, wherein, the confidence level of one-level car plate result is more than two grades of confidence levels of car plate result, two grades of cars
Confidence level of the confidence level of board result more than three-level car plate result.Multiple license plate recognition results that acquisition module is obtained are effectively
License plate recognition result.
Further, described device also includes:Second vote module, if for one-level car plate result number less than default
Value, selects two grades of car plate results from queue;If the number of one-level car plate result and two grades of summations of the number of car plate result
More than or equal to preset value, car plate ballot is carried out in all of one-level car plate result and two grades of car plate results, obtain car plate ballot
As a result.
Further, described device also includes:3rd vote module, if number and two grades of cars for one-level car plate result
The summation of the number of board result is less than preset value, and car plate ballot is carried out in all of license plate recognition result in queue, obtains
Car plate voting results.
Further, described device also includes:Whether replacement module, the Chinese character for judging in car plate voting results meets
Identification is required, if it is not, the Chinese character in car plate voting results is substituted for into default character.
In this way, the device first chooses first-class result, if poll foot first effective car plate point Three Estate when ballot
It is enough just to complete to vote.If poll not enough chooses first-class and second-class result again, if poll enough if complete ballot.If poll
Still it is insufficient to just be voted using whole results.Finally Chinese character result is judged whether by poll again need to use acquiescence
Character is replaced.The method can effectively solve ballot caused by scene causes recognition result confidence level not high due to illumination reason
The problem of error, and improve the discrimination of Chinese character.
A kind of improved car plate voting method provided by the present invention and device are described in detail above.Herein
Apply specific case to be set forth principle of the invention and implementation method, the explanation of above example is only intended to help
Understand the method for the present invention and its core concept.It should be pointed out that for those skilled in the art, not taking off
On the premise of the principle of the invention, some improvement and modification can also be carried out to the present invention, these are improved and modification also falls into this
In invention scope of the claims.
Claims (10)
1. a kind of improved car plate voting method, it is characterised in that including:
Obtain multiple license plate recognition results;
The multiple license plate recognition result is carried out into grade distinguishing, is divided into one-level car plate result, two grades of car plate results and three-level car
Board result, and the multiple license plate recognition result is added in queue;
One-level car plate result is selected from queue, if the number of one-level car plate result is more than or equal to preset value, all of one
Car plate ballot is carried out in level car plate result, car plate voting results are obtained.
2. the method for claim 1, it is characterised in that also include:
If the number of one-level car plate result is less than preset value, two grades of car plate results are selected from queue;If one-level car plate result
Number and two grades of summations of the number of car plate result be more than or equal to preset value, in all of one-level car plate result and two grades of car plates
Car plate ballot is carried out in result, car plate voting results are obtained.
3. method as claimed in claim 2, it is characterised in that also include:
If the number of one-level car plate result and two grades of summations of the number of car plate result are less than preset value, all of in queue
Car plate ballot is carried out in license plate recognition result, car plate voting results are obtained.
4. the method for claim 1, it is characterised in that the confidence level of one-level car plate result is more than two grades of car plate results
Confidence level, confidence level of two grades of confidence levels of car plate result more than three-level car plate result.
5. the method as described in any one in Claims 1-4, also includes:
Judge whether the Chinese character in car plate voting results meets identification and require, if it is not, the Chinese character in car plate voting results is replaced
Into default character.
6. a kind of improved car plate balloting device, it is characterised in that including:
Acquisition module, for obtaining multiple license plate recognition results;
Discriminating module, for the multiple license plate recognition result to be carried out into grade distinguishing, is divided into one-level car plate result, two grades of car plates
Result and three-level car plate result, and the multiple license plate recognition result is added in queue;
First vote module, for selecting one-level car plate result from queue, if the number of one-level car plate result is more than or equal to
Preset value, car plate ballot is carried out in all of one-level car plate result, obtains car plate voting results.
7. device as claimed in claim 6, it is characterised in that also include:
Second vote module, if being less than preset value for the number of one-level car plate result, selects two grades of car plate knots from queue
Really;If the number of one-level car plate result and two grades of summations of the number of car plate result are more than or equal to preset value, in all of one-level
Car plate ballot is carried out in car plate result and two grades of car plate results, car plate voting results are obtained.
8. device as claimed in claim 7, it is characterised in that also include:
3rd vote module, if the number and two grades of summations of the number of car plate result for one-level car plate result are less than default
Value, car plate ballot is carried out in all of license plate recognition result in queue, obtains car plate voting results.
9. device as claimed in claim 6, it is characterised in that the confidence level of one-level car plate result is more than two grades of car plate results
Confidence level, confidence level of two grades of confidence levels of car plate result more than three-level car plate result.
10. the device as described in any one in claim 6 to 9, it is characterised in that also include:
Replacement module, whether the Chinese character for judging in car plate voting results meets identification requires, if it is not, by car plate voting results
In Chinese character be substituted for default character.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611141388.2A CN106778743A (en) | 2016-12-12 | 2016-12-12 | A kind of improved car plate voting method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611141388.2A CN106778743A (en) | 2016-12-12 | 2016-12-12 | A kind of improved car plate voting method and device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106778743A true CN106778743A (en) | 2017-05-31 |
Family
ID=58880491
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611141388.2A Pending CN106778743A (en) | 2016-12-12 | 2016-12-12 | A kind of improved car plate voting method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106778743A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109948608A (en) * | 2019-02-28 | 2019-06-28 | 中山安信通机器人制造有限公司 | A kind of license plate number knows method for distinguishing, computer installation and computer readable storage medium |
CN110991442A (en) * | 2019-09-02 | 2020-04-10 | 南京感动科技有限公司 | High-accuracy highway license plate cloud identification method |
CN113688802A (en) * | 2021-10-22 | 2021-11-23 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
CN114677774A (en) * | 2022-03-30 | 2022-06-28 | 深圳市捷顺科技实业股份有限公司 | Barrier gate control method and related equipment |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2007109541A2 (en) * | 2006-03-16 | 2007-09-27 | Brubaker Curtis M | System and method for obtaining revenue through the display of hyper-relevant advertising on moving objects |
US8326037B1 (en) * | 2005-11-23 | 2012-12-04 | Matrox Electronic Systems, Ltd. | Methods and apparatus for locating an object in an image |
CN103824066A (en) * | 2014-03-18 | 2014-05-28 | 厦门翼歌软件科技有限公司 | Video stream-based license plate recognition method |
CN105117725A (en) * | 2015-07-29 | 2015-12-02 | 安徽四创电子股份有限公司 | License plate self-correcting identification method based on license plate credibility |
CN105631429A (en) * | 2015-12-29 | 2016-06-01 | 深圳市捷顺科技实业股份有限公司 | License plate voting method, apparatus, and system |
-
2016
- 2016-12-12 CN CN201611141388.2A patent/CN106778743A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8326037B1 (en) * | 2005-11-23 | 2012-12-04 | Matrox Electronic Systems, Ltd. | Methods and apparatus for locating an object in an image |
WO2007109541A2 (en) * | 2006-03-16 | 2007-09-27 | Brubaker Curtis M | System and method for obtaining revenue through the display of hyper-relevant advertising on moving objects |
CN103824066A (en) * | 2014-03-18 | 2014-05-28 | 厦门翼歌软件科技有限公司 | Video stream-based license plate recognition method |
CN105117725A (en) * | 2015-07-29 | 2015-12-02 | 安徽四创电子股份有限公司 | License plate self-correcting identification method based on license plate credibility |
CN105631429A (en) * | 2015-12-29 | 2016-06-01 | 深圳市捷顺科技实业股份有限公司 | License plate voting method, apparatus, and system |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109948608A (en) * | 2019-02-28 | 2019-06-28 | 中山安信通机器人制造有限公司 | A kind of license plate number knows method for distinguishing, computer installation and computer readable storage medium |
CN110991442A (en) * | 2019-09-02 | 2020-04-10 | 南京感动科技有限公司 | High-accuracy highway license plate cloud identification method |
CN110991442B (en) * | 2019-09-02 | 2023-09-22 | 南京感动科技有限公司 | High-precision identification method for license plate cloud of expressway |
CN113688802A (en) * | 2021-10-22 | 2021-11-23 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
CN113688802B (en) * | 2021-10-22 | 2022-04-01 | 季华实验室 | Gesture recognition method, device and equipment based on electromyographic signals and storage medium |
CN114677774A (en) * | 2022-03-30 | 2022-06-28 | 深圳市捷顺科技实业股份有限公司 | Barrier gate control method and related equipment |
CN114677774B (en) * | 2022-03-30 | 2023-10-17 | 深圳市捷顺科技实业股份有限公司 | Barrier gate control method and related equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106778743A (en) | A kind of improved car plate voting method and device | |
CN103927880B (en) | A kind of Car license recognition comparison method and device | |
CN109993056A (en) | A kind of method, server and storage medium identifying vehicle violation behavior | |
CN101373517B (en) | Method and system for recognizing license plate | |
CN107194398B (en) | Vehicle damages recognition methods and the system at position | |
CN109949579A (en) | A kind of illegal automatic auditing method that makes a dash across the red light based on deep learning | |
CN107967806A (en) | Vehicle fake-license detection method, device, readable storage medium storing program for executing and electronic equipment | |
CN205810112U (en) | Vehicle License Plate Recognition System | |
CN101187985B (en) | Method and device for classification boundary of identifying object classifier | |
CN102194320B (en) | High-definition network intelligent camera and high-definition network intelligent shooting method | |
CN103646550A (en) | Intelligent vehicle license plate recognition system | |
CN107491762A (en) | A kind of pedestrian detection method | |
CN106778648A (en) | Vehicle tracing and Vehicle License Plate Recognition System and recognition methods | |
CN101770642A (en) | Method and system for counting number of people in car | |
CN108509912A (en) | Multipath network video stream licence plate recognition method and system | |
CN102110369A (en) | Jaywalking snapshot method and device | |
CN107730896A (en) | Licence plate recognition method and device | |
CN109657664A (en) | A kind of recognition methods, device and the electronic equipment of license plate type | |
CN105551261A (en) | False-license-plate vehicle detection method and system | |
CN108694399A (en) | Licence plate recognition method, apparatus and system | |
US10867511B2 (en) | Apparatus and method for identifying license plate tampering | |
CN106530414A (en) | Person, certificate and ticket unified secondary detection system | |
CN112528940B (en) | Training method, recognition method and device of driver behavior recognition model | |
CN108205897A (en) | A kind of car plate analysis and identification system of vehicle | |
CN109949269A (en) | A kind of detection method and device of railroad train dust cap breakage failure |
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 |
Application publication date: 20170531 |
|
RJ01 | Rejection of invention patent application after publication |