CN110349411A - A kind of licence plate recognition method and system - Google Patents

A kind of licence plate recognition method and system Download PDF

Info

Publication number
CN110349411A
CN110349411A CN201910508318.3A CN201910508318A CN110349411A CN 110349411 A CN110349411 A CN 110349411A CN 201910508318 A CN201910508318 A CN 201910508318A CN 110349411 A CN110349411 A CN 110349411A
Authority
CN
China
Prior art keywords
recognition result
cloud
local
license plate
recognition
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910508318.3A
Other languages
Chinese (zh)
Other versions
CN110349411B (en
Inventor
陈诚
杨涛
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chengdu Ebo Information Technology Co Ltd
Original Assignee
Chengdu Ebo Information Technology Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chengdu Ebo Information Technology Co Ltd filed Critical Chengdu Ebo Information Technology Co Ltd
Priority to CN201910508318.3A priority Critical patent/CN110349411B/en
Publication of CN110349411A publication Critical patent/CN110349411A/en
Application granted granted Critical
Publication of CN110349411B publication Critical patent/CN110349411B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • GPHYSICS
    • G07CHECKING-DEVICES
    • G07BTICKET-ISSUING APPARATUS; FARE-REGISTERING APPARATUS; FRANKING APPARATUS
    • G07B15/00Arrangements or apparatus for collecting fares, tolls or entrance fees at one or more control points
    • G07B15/02Arrangements or apparatus for collecting fares, tolls or entrance fees at one or more control points taking into account a variable factor such as distance or time, e.g. for passenger transport, parking systems or car rental systems
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/017Detecting movement of traffic to be counted or controlled identifying vehicles
    • G08G1/0175Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/10Protocols in which an application is distributed across nodes in the network
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00Network arrangements or protocols for supporting network services or applications
    • H04L67/01Protocols
    • H04L67/12Protocols specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks

Abstract

The invention discloses a kind of licence plate recognition method and systems, in this method, local Car license recognition first is carried out to vehicle in inlet and exit, if recognizing vehicle is unlicensed vehicle, then carry out cloud Car license recognition, the cloud recognition result in exit is matched with the local recognition result of inlet or cloud recognition result, if be matched to it is identical as a result, if to vehicle carry out regular fee clearing.The present invention is able to ascend discrimination, while reducing cost and improving recognition speed.

Description

A kind of licence plate recognition method and system
Technical field
The present invention relates to field of license plate recognition, more particularly to a kind of licence plate recognition method and system.
Background technique
It is very universal in such a way that front end hardware identifies that license plate carries out parking lot management, using dedicated identification core Piece, discrimination is high, and recognition speed is fast, and normal condition can be more than 98%, even up to 99.5% or more, has obtained market Generally approve.But front end hardware, because the reason of cost control, calculation resources are comparatively fewer, are not able to satisfy The discrimination of complex scene, such as inverted image, speed are too fast etc..In order to promote license plate recognition rate, many managing system of car parking are also The scheme being recognized using artificial or cloud is compared the result of secondary identification to mention with hardware identification result Rise license plate recognition rate.
For the identification of cloud, because there are many resource of cloud server, various complex scenes can be coped with, so that knowing Rate is not promoted obvious, therefore cloud identification can match in excellence or beauty substantially with manual identified, reach 100% discrimination.But frequency Numerous has following problem using the secondary identification in cloud:
1. influencing recognition speed.It is higher by efficiency requirements to vehicle in the occasions such as parking lot or highway, if often Vehicle requires secondary identification, executes uploading pictures, issues the processes such as recognition result, will lead to the time and occupies and extend, reduces logical Line efficiency.
2. increasing network charges.Uploading pictures, reception recognition result can expend Internet resources, wireless in 4G or NB etc. In transmission, if be all recognized when each vehicle pass-through, network charges can be dramatically increased.
3. increasing the workload and expense of cloud computing.The resource of cloud server is also limited, and most of cloud meters Calculating the cloud Car license recognition service that supplier provides is charged according to identification number.
To sum up, front end hardware identification cost is low but discrimination is relatively low, recognition speed is fast, cloud identification it is at high cost But discrimination is also high, recognition speed is slow, the Car license recognition scheme of the prior art does not have comprehensive front-end hardware identification and cloud to know Other advantage, cannot be considered in terms of cost and discrimination.
Summary of the invention
The invention mainly solves the technical problem of providing a kind of licence plate recognition method and systems, are able to ascend discrimination, Cost is reduced simultaneously and improves recognition speed.
In order to solve the above technical problems, one technical scheme adopted by the invention is that: a kind of licence plate recognition method is provided, is wrapped It includes: S1: shooting the first license plate picture for currently driving into vehicle of inlet, and local license plate is carried out to the first license plate picture Identification obtains the first local recognition result;S2: if the first local recognition result is complete license plate number, by described the Local is stored in after one local recognition result and the first license plate picture are associated;S3: if the described first local recognition result is Incomplete license plate number then carries out cloud Car license recognition to the first license plate picture and obtains the first cloud recognition result, and will First cloud recognition result is associated with the first license plate picture as the first local recognition result;S4: exit is shot Second license plate picture of current outgoing vehicles, and local Car license recognition is carried out to the second license plate picture and obtains the second local knowledge Other result;S5: if the described second local recognition result is incomplete license plate number, cloud is carried out to the second license plate picture End Car license recognition obtains the second cloud recognition result;S6: if second cloud recognition result is complete license plate number, Search whether there is the identical with the second cloud recognition result first local recognition result in locally keeping records of;S7: if deposited In the identical with the second cloud recognition result first local recognition result, then regular fee knot is carried out to the current outgoing vehicles It calculates.
Preferably, the method also includes S8: if the described second local recognition result is complete license plate number, Search whether there is the first local recognition result identical with the second local recognition result in locally keeping records of;S9: if deposited In the first local recognition result identical with the second local recognition result, then regular fee knot is carried out to the current outgoing vehicles It calculates;S10: it if there is no the first local recognition result identical with the second local recognition result, then re-starts described to institute It states the second license plate picture and carries out the step of cloud Car license recognition obtains the second cloud recognition result.
Preferably, the method also includes S11: if second cloud recognition result is incomplete license plate number, Abnormal disbursement and sattlement is carried out to the current outgoing vehicles.
Preferably, the method also includes S12: local if there is no with the second cloud recognition result identical first Recognition result then selects corresponding at least partly license plate number the first local recognition result identical with the second cloud recognition result The first license plate picture as matching picture;S13: cloud Car license recognition is carried out to the matching picture and obtains the identification of third cloud As a result;S14: search whether there is third cloud recognition result identical with the second cloud recognition result, and by the third cloud Hold recognition result associated with the first license plate picture as the first local recognition result;S15: it is identified if there is with the second cloud As a result identical third cloud recognition result then carries out regular fee clearing to the current outgoing vehicles.
Preferably, the method also includes S16: if there is no third cloud identical with the second cloud recognition result Recognition result then carries out abnormal disbursement and sattlement to the current outgoing vehicles.
In order to solve the above technical problems, another technical solution used in the present invention is: a kind of Vehicle License Plate Recognition System is provided, It include: entrance shooting module, for shooting the first license plate picture for currently driving into vehicle of inlet, and to first license plate Picture carries out local Car license recognition and obtains the first local recognition result;Entrance cloud identification module, for local described first When recognition result is incomplete license plate number, cloud Car license recognition is carried out to the first license plate picture and obtains the identification of the first cloud As a result;Entrance processing module, for when the described first local recognition result is complete license plate number, described first locally to be known Local is stored in after other result and the first license plate picture are associated, or local using first cloud recognition result as first Recognition result is associated with the first license plate picture;Outlet shooting module, second of the current outgoing vehicles for shooting exit License plate picture, and local Car license recognition is carried out to the second license plate picture and obtains the second local recognition result;Cloud is exported to know Other module, for carrying out cloud to the second license plate picture when the described second local recognition result is incomplete license plate number End Car license recognition obtains the second cloud recognition result;Processing module is exported, for being complete in second cloud recognition result License plate number when, search whether there is the first local identification knot identical with the second cloud recognition result in locally keeping records of Fruit, and when there is the first local recognition result identical with the second cloud recognition result, to the current outgoing vehicles into The clearing of row regular fee.
Preferably, the outlet processing module is also used to when the described second local recognition result is complete license plate number, It searches whether there is the first local recognition result identical with the second local recognition result in locally keeping records of, and is depositing In the first local recognition result identical with the second local recognition result, regular fee knot is carried out to the current outgoing vehicles It calculates, when there is no the first local recognition result identical with the second local recognition result, again to the second license plate picture It carries out cloud Car license recognition and obtains the second cloud recognition result.
Preferably, the outlet processing module is also used in second cloud recognition result be incomplete license plate number When, abnormal disbursement and sattlement is carried out to the current outgoing vehicles.
Preferably, the Vehicle License Plate Recognition System further includes fuzzy matching module, and the fuzzy matching module is not for depositing In the identical with the second cloud recognition result first local recognition result, selection at least partly license plate number and the second cloud are known First license plate picture corresponding to the identical first local recognition result of other result is as matching picture;The outlet cloud identification Module is also used to carry out cloud Car license recognition to the matching picture to obtain third cloud recognition result, and by the third cloud Recognition result is associated with the first license plate picture as the first local recognition result;The fuzzy matching module is also used to search It is no to there is third cloud recognition result identical with the second cloud recognition result;The outlet processing module be also used to exist with When the identical third cloud recognition result of the second cloud recognition result, regular fee clearing are carried out to the current outgoing vehicles.
Preferably, the outlet processing module is also used to that third cloud identical with the second cloud recognition result is being not present When recognition result, abnormal disbursement and sattlement is carried out to the current outgoing vehicles.
It is in contrast to the prior art, the beneficial effects of the present invention are: by carrying out to driving into vehicle and outgoing vehicles Local Car license recognition obtains the first local recognition result and the second local recognition result, only in the first local recognition result or the When two local recognition results are incomplete license plate number, cloud Car license recognition just will do it, can reduce the knowledge of cloud license plate in this way Other number, this method so as to promote discrimination, while reducing cost and raising mainly or by local Car license recognition Recognition speed.
Detailed description of the invention
Fig. 1 is the flow diagram of the licence plate recognition method of first embodiment of the invention.
Fig. 2 is the flow diagram of the licence plate recognition method of second embodiment of the invention.
Fig. 3 is the flow diagram of the licence plate recognition method of third embodiment of the invention.
Fig. 4 is the schematic illustration of the Vehicle License Plate Recognition System of one embodiment of the invention.
Fig. 5 is the schematic illustration of the Vehicle License Plate Recognition System of another embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete Site preparation description, it is clear that the described embodiments are merely a part of the embodiments of the present invention, instead of all the embodiments.It is based on Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other Embodiment shall fall within the protection scope of the present invention.
Refering to fig. 1, be first embodiment of the invention licence plate recognition method flow diagram, in the first embodiment, This method comprises:
S1: the first license plate picture for currently driving into vehicle of inlet is shot, and local vehicle is carried out to the first license plate picture Board identifies to obtain the first local recognition result.
Wherein, local Car license recognition is identified using front end hardware, and recognition speed is fast.
S2: if the first local recognition result is complete license plate number, by the first local recognition result and the first license plate Local is stored in after picture is associated.
Wherein, complete license plate number indicates that currently driving into vehicle is to have board vehicle, and complete license plate number refers to number of license plate number Number of codes and format meet preset rules, such as the license plate number of China, and river A12345 is just complete License plate number.
S3: if the first local recognition result is incomplete license plate number, cloud license plate is carried out to the first license plate picture Identification obtains the first cloud recognition result, and using the first cloud recognition result as the first local recognition result and the first license plate figure Piece is associated.
Wherein, cloud Car license recognition is identified using cloud, and discrimination is high, it is believed that the result of cloud Car license recognition is most Terminate fruit.Incomplete license plate number indicates the number quantity of license plate number and format meets preset rules, for example, river A1234 due to It is incomplete license plate number, reason may be the license plate breakage of the vehicle or be blocked, then example as soon as having lacked a number If the first local recognition result is blank, without any number, then being also incomplete license plate number, reason may be the vehicle Be new car, there are no suspension license plate.Whatsoever situation, incomplete license plate number all indicate that the vehicle is unlicensed vehicle, relatively , complete license plate number then indicates that the vehicle is to have board vehicle.
If it is local keep records of in there are no the record of the first license plate picture, using the first cloud recognition result as the Local is stored in after one local recognition result and the first license plate picture are associated.
S4: the second license plate picture of the current outgoing vehicles in exit is shot, and local vehicle is carried out to the second license plate picture Board identifies to obtain the second local recognition result.
S5: if the second local recognition result is incomplete license plate number, cloud license plate is carried out to the second license plate picture Identification obtains the second cloud recognition result.
S6: if the second cloud recognition result is complete license plate number, search whether exist in locally keeping records of Identical with the second cloud recognition result first local recognition result.
S7: if there is the identical with the second cloud recognition result first local recognition result, then to current outgoing vehicles Carry out regular fee clearing.
Wherein, if local keeps records of middle presence the first local recognition result identical with the second cloud recognition result, So show that the local Car license recognition of inlet accurately identifies, but the local license plate number wrong identification in exit, but go out Cloud Car license recognition at mouthful accurately identifies again, thus the second cloud recognition result is mutually matched with the first local recognition result, So as to according to the disbursement and sattlements such as the two corresponding time because usually carry out regular fee clearing.
In existing managing system of car parking, a maximum risk is that (new car does not hang license plate to unlicensed vehicle, license plate is hidden Gear or license plate are damaged) the problem of, unlicensed vehicle must be managed using any special measures (such as parking card).But when there is board If vehicle misrecognition is at unlicensed vehicle, the process that will lead to whole system is chaotic, for example inlet has board vehicle to be known by mistake Not Wei unlicensed vehicle, card taking admission, this has board vehicle to be accurately identified again when appearance, this will lead to the loss etc. of parking card.Using The Car license recognition scheme of single front end hardware identification not can avoid the problem then, the Car license recognition scheme identified using single cloud Although can be to avoid the problem, cloud identification is all carried out for each car will definitely increase cost, reduce recognition speed and Traffic efficiency.And the method for the present embodiment just will start cloud vehicle when it is unlicensed vehicle that local Car license recognition is by vehicle identification Board identification, can reduce the number of cloud Car license recognition as far as possible in this way, so as to promote discrimination, while reduce as far as possible Cost and raising recognition speed.
The method that the present embodiment is exemplified below, in instances, the license plate number of certain trolley (calling A vehicle in the following text) are river A12345。
Example 1:A vehicle obtains the first local recognition result by local Car license recognition in inlet, is incomplete vehicle The trade mark, and then continue cloud Car license recognition and obtain the first cloud recognition result, it is river A12345, the first cloud is identified As a result as the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, A vehicle arrives Up to exit, it is incomplete license plate that A vehicle, which obtains the second local recognition result by local Car license recognition in exit, Number, and then continue cloud Car license recognition and obtain the second cloud recognition result, it is river A12345, in locally keeping records of The first local recognition result of river A12345 is found, it is identical as the second cloud recognition result, illustrate that current outgoing vehicles are A Vehicle, and then regular fee clearing are carried out according to the first local recognition result and the second cloud recognition result.
Example 2:A vehicle, which obtains the first local recognition result river A12345 by local Car license recognition in inlet, (accurately to be known Not), by the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, A vehicle reaches Exit, A vehicle obtain the second local recognition result by local Car license recognition in exit, are incomplete license plate number, And then continue cloud Car license recognition and obtain the second cloud recognition result, it is river A12345, is looked into locally keeping records of The first local recognition result of river A12345 is found, it is identical as the second cloud recognition result, illustrate that current outgoing vehicles are A vehicle , and then regular fee clearing are carried out according to the first local recognition result and the second cloud recognition result.
Referring to Fig.2, be the flow diagram of the licence plate recognition method of second embodiment of the invention, in a second embodiment, This method contains the technical characteristic with first embodiment whole, the difference is that, this method further include:
S8: if the second local recognition result is complete license plate number, search whether exist in locally keeping records of The first local recognition result identical with the second local recognition result.
S9: if there is the first local recognition result identical with the second local recognition result, then to current outgoing vehicles Carry out regular fee clearing.
S10: it if there is no the first local recognition result identical with the second local recognition result, then re-starts to the Two license plate pictures carry out the step of cloud Car license recognition obtains the second cloud recognition result.
Further, in the present embodiment, this method further include:
S11: if the second cloud recognition result is incomplete license plate number, abnormal expense is carried out to current outgoing vehicles Clearing.
Wherein, abnormal disbursement and sattlement can be is settled accounts according to preset strategy, be also possible to issue prompt by manually into Row clearing.
Still illustrate the method for the present embodiment by taking A vehicle as an example below.
Example 1:A vehicle obtains the first local recognition result by local Car license recognition in inlet, is incomplete vehicle The trade mark, and then continue cloud Car license recognition and obtain the first cloud recognition result river A12345, by the first cloud recognition result As the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, A vehicle reaches out At mouthful, A vehicle obtains the second local recognition result river A12346 (wrong identification) by local Car license recognition in exit, at this Ground keeps records of middle the first local recognition result searched less than river A12346, and then continues cloud Car license recognition and obtain the Two cloud recognition result river A12345 find the first local recognition result of river A12345 in locally keeping records of, with Second cloud recognition result is identical, illustrates that current outgoing vehicles are A vehicle, and then according to the first local recognition result and the second cloud Recognition result is held to carry out regular fee clearing.
Example 2:A vehicle obtains the first local recognition result by local Car license recognition in inlet, is incomplete vehicle The trade mark, and then continue cloud Car license recognition and obtain the first cloud recognition result river A12345, by the first cloud recognition result As the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, A vehicle reaches out At mouthful, A vehicle obtains the second local recognition result river A12345 (accurately identifying) by local Car license recognition in exit, at this Ground finds the first local recognition result of river A12345 in keeping records of, and illustrates that current outgoing vehicles are A vehicle, and then basis First local recognition result and the second local recognition result carry out regular fee clearing.
Example 3:A vehicle obtains the first local recognition result by local Car license recognition in inlet, is incomplete vehicle The trade mark, and then continue cloud Car license recognition and obtain the first cloud recognition result, it is river A12345, the first cloud is identified As a result as the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, there is vehicle (may be A vehicle, it is also possible to other vehicles) reaches exit, which obtains the by local Car license recognition in exit Two local recognition results, are incomplete license plate number, and then continue cloud Car license recognition and obtain the second cloud identification knot Fruit, but it is also incomplete license plate number, shows that current outgoing vehicles are strictly unlicensed vehicle, and then is carried out to current outgoing vehicles Abnormal disbursement and sattlement.
It is the flow diagram of the licence plate recognition method of third embodiment of the invention refering to Fig. 3, in the third embodiment, This method contains the technical characteristic with the first embodiment or the second embodiment whole, the difference is that, this method further include:
S12: it if there is no the identical with the second cloud recognition result first local recognition result, then selects at least partly First license plate picture corresponding to license plate number the first local recognition result identical with the second cloud recognition result, which is used as, to be matched Picture.
Wherein, at least partly license plate number is the fixed license plate number in position, such as first 3.
S13: cloud Car license recognition is carried out to matching picture and obtains third cloud recognition result.
S14: search whether there is third cloud recognition result identical with the second cloud recognition result, and by third cloud Recognition result is associated with the first license plate picture as the first local recognition result.
S15: if there is third cloud recognition result identical with the second cloud recognition result, then to current outgoing vehicles Carry out regular fee clearing.
Further, in the present embodiment, this method further include:
S16: if there is no third cloud recognition result identical with the second cloud recognition result, then vehicle is driven out to current Carry out abnormal disbursement and sattlement.
Still illustrate the method for the present embodiment by taking A vehicle as an example below.
Example 1:A vehicle obtains the first local recognition result river A12347 (mistake knowledge by local Car license recognition in inlet Not), by the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, A vehicle reaches Exit, A vehicle obtain the second local recognition result river A12346 (wrong identification) by local Car license recognition in exit, Middle the first local recognition result searched less than river A12346 is locally kept records of, and then continues cloud Car license recognition and obtains Second cloud recognition result river A12345, but the first local identification less than river A12345 is also searched in locally keeping records of As a result, selecting the first recognition result at this time is respectively the first license plate figure corresponding to river A12347, river A12349 and river A12348 Piece in these first recognition results, removes " river A " outside, before preceding 3 license plate numbers and the second cloud recognition result river A12345 3 license plate numbers are identical.Cloud Car license recognition is carried out to these the first license plate pictures and obtains third cloud recognition result, by third Cloud recognition result as the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.If A If first license plate picture of vehicle is erroneously identified in inlet, then the picture is after the Car license recognition of cloud, third cloud End recognition result is river A12345, identical as the second cloud recognition result, illustrates that current outgoing vehicles are A vehicle, Jin Ergen Regular fee clearing are carried out according to the first local recognition result and the second cloud recognition result.
Example 2:A vehicle, which obtains the first local recognition result river A12345 by local Car license recognition in inlet, (accurately to be known Not), by the first local recognition result it is associated with the first license plate picture of A vehicle after be stored in local.Later, which reaches Exit, the vehicle obtain the second local recognition result river A12346 (wrong identification) by local Car license recognition in exit, The first local recognition result less than river A12346 is searched in locally keeping records of, and then is continued cloud Car license recognition and obtained To the second cloud recognition result river A12345, the first local recognition result of river A12345 is found in locally keeping records of, It is identical as the second cloud recognition result, illustrates that current outgoing vehicles be license plate number are A vehicle, and then local identifies according to first As a result regular fee clearing are carried out with the second cloud recognition result.
By the above-mentioned means, the licence plate recognition method of the present embodiment is by carrying out local vehicle to driving into vehicle and outgoing vehicles Board identifies to obtain the first local recognition result and the second local recognition result, only local in the first local recognition result or second When recognition result is that incomplete license plate number or the first local recognition result are unmatched with the second local recognition result, It will do it cloud Car license recognition, the number of cloud Car license recognition can be reduced in this way, this method is mainly or by local license plate Identification so as to promote discrimination, while reducing cost and improving recognition speed.
Referring to fig. 4, be one embodiment of the invention Vehicle License Plate Recognition System schematic illustration.The system includes entrance shooting Module 11, entrance cloud identification module 12, entrance processing module 13, outlet shooting module 14, outlet 15 and of cloud identification module Export processing module 16.
Entrance shooting module 11 is used to shoot the first license plate picture for currently driving into vehicle of inlet, and to the first license plate Picture carries out local Car license recognition and obtains the first local recognition result.Wherein, local Car license recognition is identified using front end hardware, is known Other speed is fast.
Entrance cloud identification module 12 is used for when the first local recognition result is incomplete license plate number, to the first license plate Picture carries out cloud Car license recognition and obtains the first cloud recognition result.Cloud Car license recognition identifies that discrimination is high using cloud, can To think the result of cloud Car license recognition as final result.Incomplete license plate number indicates the number quantity and format of license plate number Meet preset rules, as soon as such as river A1234 due to having lacked a number, be incomplete license plate number, reason may be this The license plate of vehicle is damaged or is blocked, then the such as first local recognition result is blank, without any number, then also for not Complete license plate number, reason may be that the vehicle is new car, and there are no suspension license plates.
Entrance processing module 13 is used for when the first local recognition result is complete license plate number, by the first local identification knot Local is stored in after fruit and the first license plate picture are associated, or using the first cloud recognition result as the first local recognition result It is associated with the first license plate picture.Wherein, complete license plate number indicates that currently driving into vehicle is to have board vehicle, and complete license plate number is The number quantity and format for referring to license plate number meet preset rules, such as the license plate number of China, river A12345 is just complete license plate number.If in locally keeping records of, there are no the records of the first license plate picture, by the first cloud End recognition result is stored in local after being associated as the first local recognition result and the first license plate picture.
Outlet shooting module 14 is used to shoot the second license plate picture of the current outgoing vehicles in exit, and to the second license plate Picture carries out local Car license recognition and obtains the second local recognition result.
Cloud identification module 15 is exported to be used for when the second local recognition result is incomplete license plate number, to the second license plate Picture carries out cloud Car license recognition and obtains the second cloud recognition result.
It exports processing module 16 to be used for when the second cloud recognition result is complete license plate number, in locally keeping records of It searches whether there is the identical with the second cloud recognition result first local recognition result, and is identified existing with the second cloud As a result when the identical first local recognition result, regular fee clearing are carried out to the current outgoing vehicles.Wherein, if it is local Middle presence the first local recognition result identical with the second cloud recognition result is kept records of, then showing the local vehicle of inlet Board identification accurately identifies, but the local license plate number wrong identification in exit, but the cloud Car license recognition in exit is again quasi- Really identification, thus the second cloud recognition result is mutually matched with the first local recognition result, so as to right respectively according to the two The disbursement and sattlements such as the time answered are because usually carrying out regular fee clearing.
In existing managing system of car parking, a maximum risk is that (new car does not hang license plate to unlicensed vehicle, license plate is hidden Gear or license plate are damaged) the problem of, unlicensed vehicle must be managed using any special measures (such as parking card).But when there is board If vehicle misrecognition is at unlicensed vehicle, the process that will lead to whole system is chaotic, for example inlet has board vehicle to be known by mistake Not Wei unlicensed vehicle, card taking admission, this has board vehicle to be accurately identified again when appearance, this will lead to the loss etc. of parking card.Using The Car license recognition scheme of single front end hardware identification not can avoid the problem then, the Car license recognition scheme identified using single cloud Although can be to avoid the problem, cloud identification is all carried out for each car will definitely increase cost, reduce recognition speed and Traffic efficiency.And the system of the present embodiment just will start cloud vehicle when it is unlicensed vehicle that local Car license recognition is by vehicle identification Board identification, can reduce the number of cloud Car license recognition as far as possible in this way, so as to promote discrimination, while reduce as far as possible Cost and raising recognition speed.
Please continue to refer to Fig. 2, in another embodiment, Vehicle License Plate Recognition System contains the Car license recognition system with previous embodiment The all technical features of system, the difference is that, outlet processing module 16 is also used in the second local recognition result be complete When license plate number, search whether that there is the first local identification identical with the second local recognition result ties in locally keeping records of Fruit, and when there is the first local recognition result identical with the second local recognition result, current outgoing vehicles are carried out just Normal disbursement and sattlement, when there is no the first local recognition result identical with the second local recognition result, again to the second license plate Picture carries out cloud Car license recognition and obtains the second cloud recognition result.
Further, in the present embodiment, outlet processing module 16 is also used in the second cloud recognition result be imperfect License plate number when, abnormal disbursement and sattlement is carried out to current outgoing vehicles.
It is the schematic illustration of the Vehicle License Plate Recognition System of another embodiment of the present invention refering to Fig. 5.In the present embodiment, vehicle Board identifying system contains all technical features with aforementioned any Vehicle License Plate Recognition System, the difference is that, Car license recognition System further includes fuzzy matching module 17, and fuzzy matching module 17 is used for there is no identical with the second cloud recognition result the When one local recognition result, selection at least partly license plate number the first local recognition result identical with the second cloud recognition result The first corresponding license plate picture is as matching picture;Outlet cloud identification module 15 is also used to carry out cloud vehicle to matching picture Board identifies to obtain third cloud recognition result, and using third cloud recognition result as the first local recognition result and the first license plate Picture is associated;Fuzzy matching module 17, which is also used to search whether to have third cloud identical with the second cloud recognition result, to be known Other result;Outlet processing module 16 is also used to when there is third cloud recognition result identical with the second cloud recognition result, Regular fee clearing are carried out to current outgoing vehicles.
Further, in the present embodiment, outlet processing module 16 is also used to be not present and the second cloud recognition result When identical third cloud recognition result, abnormal disbursement and sattlement is carried out to current outgoing vehicles.
By the above-mentioned means, the Vehicle License Plate Recognition System of the present embodiment is by carrying out local vehicle to driving into vehicle and outgoing vehicles Board identifies to obtain the first local recognition result and the second local recognition result, only local in the first local recognition result or second When recognition result is that incomplete license plate number or the first local recognition result are unmatched with the second local recognition result, It will do it cloud Car license recognition, the number of cloud Car license recognition can be reduced in this way, this method is mainly or by local license plate Identification so as to promote discrimination, while reducing cost and improving recognition speed.
The above description is only an embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills Art field, is included within the scope of the present invention.

Claims (10)

1. a kind of licence plate recognition method characterized by comprising
S1: the first license plate picture for currently driving into vehicle of inlet is shot, and local vehicle is carried out to the first license plate picture Board identifies to obtain the first local recognition result;
S2: if the described first local recognition result is complete license plate number, by the described first local recognition result and first Local is stored in after license plate picture is associated;
S3: if the described first local recognition result is incomplete license plate number, cloud is carried out to the first license plate picture Car license recognition obtains the first cloud recognition result, and using first cloud recognition result as the first local recognition result and the One license plate picture is associated;
S4: the second license plate picture of the current outgoing vehicles in exit is shot, and local vehicle is carried out to the second license plate picture Board identifies to obtain the second local recognition result;
S5: if the described second local recognition result is incomplete license plate number, cloud is carried out to the second license plate picture Car license recognition obtains the second cloud recognition result;
S6: if second cloud recognition result is complete license plate number, search whether exist in locally keeping records of Identical with the second cloud recognition result first local recognition result;
S7: if there is the identical with the second cloud recognition result first local recognition result, then to the current outgoing vehicles Carry out regular fee clearing.
2. licence plate recognition method according to claim 1, which is characterized in that the method also includes:
S8: if the described second local recognition result is complete license plate number, search whether exist in locally keeping records of The first local recognition result identical with the second local recognition result;
S9: if there is the first local recognition result identical with the second local recognition result, then to the current outgoing vehicles Carry out regular fee clearing;
S10: it if there is no the first local recognition result identical with the second local recognition result, then re-starts described to institute It states the second license plate picture and carries out the step of cloud Car license recognition obtains the second cloud recognition result.
3. licence plate recognition method according to claim 1, which is characterized in that the method also includes:
S11: if second cloud recognition result is incomplete license plate number, the current outgoing vehicles are carried out abnormal Disbursement and sattlement.
4. licence plate recognition method according to any one of claims 1 to 3, which is characterized in that the method also includes:
S12: if there is no the identical with the second cloud recognition result first local recognition result, then at least partly license plate is selected First license plate picture corresponding to number the first local recognition result identical with the second cloud recognition result, which is used as, matches picture;
S13: cloud Car license recognition is carried out to the matching picture and obtains third cloud recognition result;
S14: search whether there is third cloud recognition result identical with the second cloud recognition result, and by the third cloud Recognition result is associated with the first license plate picture as the first local recognition result;
S15: if there is third cloud recognition result identical with the second cloud recognition result, then to the current outgoing vehicles Carry out regular fee clearing.
5. licence plate recognition method according to claim 4, which is characterized in that the method also includes:
S16: if there is no third cloud recognition result identical with the second cloud recognition result, then vehicle is currently driven out to described Carry out abnormal disbursement and sattlement.
6. a kind of Vehicle License Plate Recognition System characterized by comprising
Entrance shooting module, for shooting the first license plate picture for currently driving into vehicle of inlet, and to first license plate Picture carries out local Car license recognition and obtains the first local recognition result;
Entrance cloud identification module is used for when the described first local recognition result is incomplete license plate number, to described first License plate picture carries out cloud Car license recognition and obtains the first cloud recognition result;
Entrance processing module, for when the described first local recognition result is complete license plate number, described first locally to be known Local is stored in after other result and the first license plate picture are associated, or local using first cloud recognition result as first Recognition result is associated with the first license plate picture;
Outlet shooting module, the second license plate picture of the current outgoing vehicles for shooting exit, and to second license plate Picture carries out local Car license recognition and obtains the second local recognition result;
Cloud identification module is exported, is used for when the described second local recognition result is incomplete license plate number, to described second License plate picture carries out cloud Car license recognition and obtains the second cloud recognition result;
Processing module is exported, is used for when second cloud recognition result is complete license plate number, in locally keeping records of It searches whether there is the identical with the second cloud recognition result first local recognition result, and is identified existing with the second cloud As a result when the identical first local recognition result, regular fee clearing are carried out to the current outgoing vehicles.
7. Vehicle License Plate Recognition System according to claim 6, which is characterized in that the outlet processing module is also used to described When second local recognition result is complete license plate number, searches whether to exist in locally keeping records of and be tied with the second local identification The identical first local recognition result of fruit, and there is the first local recognition result identical with the second local recognition result When, regular fee clearing are carried out to the current outgoing vehicles, first identical with the second local recognition result is being not present When ground recognition result, cloud Car license recognition is carried out to the second license plate picture again and obtains the second cloud recognition result.
8. Vehicle License Plate Recognition System according to claim 6, which is characterized in that the outlet processing module is also used to described When second cloud recognition result is incomplete license plate number, abnormal disbursement and sattlement is carried out to the current outgoing vehicles.
9. according to the described in any item Vehicle License Plate Recognition Systems of claim 6 to 8, which is characterized in that the Vehicle License Plate Recognition System is also Including fuzzy matching module, the fuzzy matching module is used for local there is no identical with the second cloud recognition result first When recognition result, corresponding to selection at least partly license plate number the first local recognition result identical with the second cloud recognition result The first license plate picture as matching picture;
Outlet cloud identification module is also used to carry out the matching picture cloud Car license recognition to obtain the identification of third cloud As a result, and third cloud recognition result is associated with the first license plate picture as the first local recognition result;
The fuzzy matching module is also used to search whether to have third cloud identical with the second cloud recognition result identification knot Fruit;
The outlet processing module is also used to when there is third cloud recognition result identical with the second cloud recognition result, right The current outgoing vehicles carry out regular fee clearing.
10. Vehicle License Plate Recognition System according to claim 9, which is characterized in that the outlet processing module is also used to not When in the presence of third cloud recognition result identical with the second cloud recognition result, abnormal expense is carried out to the current outgoing vehicles Clearing.
CN201910508318.3A 2019-06-13 2019-06-13 License plate recognition method and system Active CN110349411B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910508318.3A CN110349411B (en) 2019-06-13 2019-06-13 License plate recognition method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910508318.3A CN110349411B (en) 2019-06-13 2019-06-13 License plate recognition method and system

Publications (2)

Publication Number Publication Date
CN110349411A true CN110349411A (en) 2019-10-18
CN110349411B CN110349411B (en) 2021-05-25

Family

ID=68181856

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910508318.3A Active CN110349411B (en) 2019-06-13 2019-06-13 License plate recognition method and system

Country Status (1)

Country Link
CN (1) CN110349411B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113284349A (en) * 2021-07-23 2021-08-20 深圳市赛菲姆科技有限公司 License plate recognition method and system
CN114023076A (en) * 2021-10-22 2022-02-08 北京商海文天科技发展有限公司 Specific vehicle tracking method based on multi-source heterogeneous data
US20220292836A1 (en) * 2021-03-10 2022-09-15 FlashParking, Inc. Method and system for vehicle authentication

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104978768A (en) * 2015-05-26 2015-10-14 蒋君伟 Enclosed parking lot gateway intelligent fee charging and collecting system
CN106205199A (en) * 2016-08-31 2016-12-07 广东艾科智泊科技股份有限公司 A kind of car plate fault-tolerance processing system
WO2017004760A1 (en) * 2015-07-04 2017-01-12 冯旋宇 Remote query system and method for status information of parking lot
CN106469506A (en) * 2015-08-19 2017-03-01 西安全星智能电子有限责任公司 Parking lot control method for vehicle and device
CN106558115A (en) * 2015-09-28 2017-04-05 西安艾润物联网技术服务有限责任公司 Parking lot vehicles while passing control method and device
CN107680382A (en) * 2017-02-24 2018-02-09 上海享泊信息科技有限公司 Licence plate recognition method, module, cloud server and managing system of car parking
CN109147068A (en) * 2018-09-12 2019-01-04 西安艾润物联网技术服务有限责任公司 Unlicensed vehicles management method, device and computer readable storage medium

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104978768A (en) * 2015-05-26 2015-10-14 蒋君伟 Enclosed parking lot gateway intelligent fee charging and collecting system
WO2017004760A1 (en) * 2015-07-04 2017-01-12 冯旋宇 Remote query system and method for status information of parking lot
CN106469506A (en) * 2015-08-19 2017-03-01 西安全星智能电子有限责任公司 Parking lot control method for vehicle and device
CN106558115A (en) * 2015-09-28 2017-04-05 西安艾润物联网技术服务有限责任公司 Parking lot vehicles while passing control method and device
CN106205199A (en) * 2016-08-31 2016-12-07 广东艾科智泊科技股份有限公司 A kind of car plate fault-tolerance processing system
CN107680382A (en) * 2017-02-24 2018-02-09 上海享泊信息科技有限公司 Licence plate recognition method, module, cloud server and managing system of car parking
CN109147068A (en) * 2018-09-12 2019-01-04 西安艾润物联网技术服务有限责任公司 Unlicensed vehicles management method, device and computer readable storage medium

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20220292836A1 (en) * 2021-03-10 2022-09-15 FlashParking, Inc. Method and system for vehicle authentication
CN113284349A (en) * 2021-07-23 2021-08-20 深圳市赛菲姆科技有限公司 License plate recognition method and system
CN114023076A (en) * 2021-10-22 2022-02-08 北京商海文天科技发展有限公司 Specific vehicle tracking method based on multi-source heterogeneous data
CN114023076B (en) * 2021-10-22 2022-10-04 北京商海文天科技发展有限公司 Specific vehicle tracking method based on multi-source heterogeneous data

Also Published As

Publication number Publication date
CN110349411B (en) 2021-05-25

Similar Documents

Publication Publication Date Title
CN110349411A (en) A kind of licence plate recognition method and system
CN106971557B (en) A kind of vehicle identification method and system
CN101527082B (en) System and method for acquiring traffic road conditions
CN105702047B (en) Car license recognition mistake filter method and device in a kind of deck analysis
CN106781515A (en) A kind of vehicle cloud identifying system and vehicle cloud management system
CN106781669A (en) Recommend method and device in a kind of parking stall
CN109035447A (en) A kind of Logistics Park vehicle parking charging method and system
CN107331170B (en) Intersection signal timing evaluation method and system based on the green light loss time
CN112700641B (en) Parking lot management method and device, electronic equipment and storage medium
CN106355656B (en) A kind of implementation method of intelligent parking lot, device and system
CN106710292B (en) Large parking lot vehicles management method and system
CN105551288A (en) Intelligent parking lot traffic management system based on internet of things
CN111724584A (en) Bus signal priority high-frequency multi-application-oriented dynamic control method
CN109658535A (en) A kind of attendance processing method and processing device based on vehicle
CN110047319A (en) Parking position air navigation aid, electronic device and storage medium
CN112382102A (en) Intelligent control system of urban traffic signal lamp
CN109508404A (en) Repair instructional video management method, device, terminal device and storage medium
CN108898841A (en) A kind of processing method and processing device of parking lot vehicle identification exception
CN111243275A (en) Toll station traffic state monitoring device and method based on computer vision
CN108805309A (en) A kind of parking system and its workflow based on block chain technology
CN115641359B (en) Method, device, electronic equipment and medium for determining movement track of object
CN109637187A (en) City Roadside Parking position unmanned charge monitoring and managing method and system
CN107221165A (en) Parking lot access management system
CN112382103A (en) Intelligent control method and storage medium for urban traffic signal lamp
CN106878031B (en) Method and system for accelerating processing of telephone bill of telecommunication package user

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
GR01 Patent grant
GR01 Patent grant