CN110084232A - The recognition methods of chinese character, device and terminal device in license plate - Google Patents
The recognition methods of chinese character, device and terminal device in license plate Download PDFInfo
- Publication number
- CN110084232A CN110084232A CN201810071711.6A CN201810071711A CN110084232A CN 110084232 A CN110084232 A CN 110084232A CN 201810071711 A CN201810071711 A CN 201810071711A CN 110084232 A CN110084232 A CN 110084232A
- Authority
- CN
- China
- Prior art keywords
- chinese character
- character
- license plate
- goal
- selling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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 present invention relates to technical field of vehicle, more particularly to recognition methods, device and the terminal device of chinese character in a kind of license plate, the recognition methods of chinese character and device are applied to terminal device in the license plate, the recognition methods of chinese character includes: reception license plate image in the license plate, and then calculate the confidence level of each chinese character in the license plate image, if the confidence level of the chinese character is lower than preset matching threshold value, goal-selling character is set by the chinese character.Wherein, if the confidence level of the chinese character is lower than preset matching threshold value, show the recognition result inaccuracy of the chinese character, therefore, set goal-selling character for the chinese character, to improve the accuracy rate of chinese character identification.It should be noted that the preset matching threshold value and goal-selling character are all real-time updates, therefore, by that can greatly improve the accuracy rate of chinese character in license plate using the data as standard.
Description
Technical field
The present invention relates to technical field of vehicle, in particular to the recognition methods of chinese character, device in a kind of license plate
And terminal device.
Background technique
In intelligent transportation road management system, Car license recognition is the technology of wherein core, and the accuracy of Car license recognition is direct
It is related to the validity of road monitoring and law enforcement.But for domestic license plate, there are the Chinese characters of stroke complexity, unlike Europe
License plate only has number and letter, because the identification difficulty of Chinese character is relatively high, domestic Car license recognition requires higher.Therefore,
The algorithm of domestic manufacturers can all have the configuration item in province, and all can be straight in the case that Chinese character confidence level is not high under fuzzy scene
Connect the province assigned into default configuration.Such method is under the more scene of external vehicle and is not suitable for, such as high speed, municipality directly under the Central Government
Equal regions are not very effectively, to the recognition accuracy that promotes Chinese character and its limited.And there is such configuration item to exist, also increase
The complexity of configuration is added.Therefore it provides a kind of recognition accuracy for improving chinese character in license plate is highly important.
Summary of the invention
The purpose of the present invention is to provide a kind of recognition methods of chinese character in license plate, to improve chinese character identification
Accuracy rate.
Another object of the present invention is to provide a kind of identification devices of chinese character in license plate, to improve chinese character knowledge
Other accuracy rate.
Another object of the present invention is to provide a kind of terminal devices, to improve the accuracy rate of chinese character identification.
To achieve the goals above, technical solution used in the embodiment of the present invention is as follows:
In a first aspect, the embodiment of the invention provides a kind of recognition methods of chinese character in license plate, which comprises
Receive license plate image；Calculate the confidence level of chinese character in the license plate image；If the confidence level of the chinese character is lower than pre-
If matching threshold, then goal-selling character, the preset matching threshold value and goal-selling character are set by the chinese character
It is that real-time calculate updates.
Second aspect, the embodiment of the invention also provides a kind of identification device of chinese character in license plate, described device packets
It includes: receiving module, for receiving license plate image；Computing module calculates the confidence level of chinese character in the license plate image；Setting
Module sets default mesh for man's character if the confidence level for the chinese character is lower than preset matching threshold value
Marking-up symbol, the preset matching threshold value and goal-selling character are that real-time calculate updates.
The third aspect, the embodiment of the invention also provides a kind of terminal device, the terminal device includes memory, processing
Device, the memory is for storing computer program code, and the processor is based on executing and being stored in the memory
Calculation machine program code is to realize the recognition methods of chinese character in the license plate.
Recognition methods, device and the terminal device of chinese character, the license plate in a kind of license plate provided in an embodiment of the present invention
The recognition methods of middle chinese character and device are applied to terminal device, and the recognition methods of chinese character includes: to receive in the license plate
License plate image, and then the confidence level of each chinese character in the license plate image is calculated, if the confidence level of the chinese character is lower than
Preset matching threshold value then sets goal-selling character for the chinese character.Wherein, if the confidence level of the chinese character is lower than pre-
If matching threshold, shows the recognition result inaccuracy of the chinese character, therefore, set goal-selling word for the chinese character
Symbol, to improve the accuracy rate of chinese character identification.It should be noted that the preset matching threshold value and goal-selling character are all
Real-time update, therefore, by the way that the accuracy rate of chinese character in license plate can be greatlyd improve using the data as standard.
To enable the above objects, features and advantages of the present invention to be clearer and more comprehensible, preferred embodiment is cited below particularly, and cooperate
Appended attached drawing, is described in detail below.
Detailed description of the invention
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be understood that the following drawings illustrates only certain embodiments of the present invention, therefore is not construed as pair
The restriction of range for those of ordinary skill in the art without creative efforts, can also be according to this
A little attached drawings obtain other relevant attached drawings.
Fig. 1 shows a kind of structural schematic diagram of terminal device provided in an embodiment of the present invention.
Fig. 2 shows the flow diagrams of the recognition methods of chinese character in a kind of license plate provided in an embodiment of the present invention.
Fig. 3 shows a kind of flow diagram of the calculation method of goal-selling character provided in an embodiment of the present invention.
Fig. 4 shows a kind of flow diagram of the calculation method of preset matching threshold value provided in an embodiment of the present invention.
Fig. 5 shows the functional module signal of the identification device of chinese character in a kind of license plate provided in an embodiment of the present invention
Figure.
Diagram: 100- terminal device；The identification device of chinese character in 110- license plate；120- memory；130- storage control
Device processed；140- processor；150- Peripheral Interface；160- input-output unit；170- audio unit；180- display unit；190-
Communication unit；111- receiving module；112- computing module；113- chooses module；114- statistical module；115- setup module.
Specific embodiment
Below in conjunction with attached drawing in the embodiment of the present invention, technical solution in the embodiment of the present invention carries out clear, complete
Ground description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Usually exist
The component of the embodiment of the present invention described and illustrated in attached drawing can be arranged and be designed with a variety of different configurations herein.Cause
This, is not intended to limit claimed invention to the detailed description of the embodiment of the present invention provided in the accompanying drawings below
Range, but it is merely representative of selected embodiment of the invention.Based on the embodiment of the present invention, those skilled in the art are not doing
Every other embodiment obtained under the premise of creative work out, shall fall within the protection scope of the present invention.
It should also be noted that similar label and letter indicate similar terms in following attached drawing, therefore, once a certain Xiang Yi
It is defined in a attached drawing, does not then need that it is further defined and explained in subsequent attached drawing.Meanwhile of the invention
In description, term " first ", " second " etc. are only used for distinguishing description, are not understood to indicate or imply relative importance.
Car license recognition is of great significance to road monitoring and law enforcement, since Chinese character is extremely complex, as a result, just to Chinese character
Really identification is more difficult.In addition, in urban road monitor, scene and its complexity, such as interference of low visibility, various environment light
Deng increasing the difficulty to Chinese Character Recognition more.The embodiment of the present invention provides a kind of identification side of chinese character in license plate as a result,
Method, this method adjust parameter preset by dynamic, so that detection means has better real-time and accuracy.This method application
In terminal device, i other words, which carries out the image that the camera being set under each scene to distribution acquires
Analysis, and parameter preset of the adjustment as standard in real time, to improve the accuracy that chinese character identifies in license plate.
The terminal device may be, but not limited to, and the intelligent electronic devices such as desktop computer please refer to Fig. 1, be of the invention real
A kind of structural schematic diagram of the terminal device 100 provided in example is applied, which includes the identification of chinese character in license plate
Device 110, memory 120, storage control 130, processor 140, Peripheral Interface 150, input-output unit 160, audio list
Member 170, display unit 180 and communication unit 190.
The memory 120, storage control 130, processor 140, Peripheral Interface 150, input-output unit 160, sound
Frequency unit 170, display unit 180, each element of communication unit 190 are directly or indirectly electrically connected between each other, to realize number
According to transmission or interaction.For example, these elements can be realized electrically between each other by one or more communication bus or signal wire
Connection.The identification device 110 of chinese character can be with the shape of software or firmware (firmware) including at least one in the license plate
Formula is stored in the memory 120 or is solidificated in the operating system (operating of 100 equipment of terminal device
System, OS) in software function module.The processor 140 is used to execute the executable module stored in memory 120,
Such as the identification device 110 of chinese character includes in the license plate software function module or computer program.
Wherein, memory 120 may be, but not limited to, random access memory (Random Access Memory,
RAM), read-only memory (Read Only Memory, ROM), programmable read only memory (Programmable Read-Only
Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only Memory, EPROM),
Electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only Memory, EEPROM) etc..
Wherein, memory 120 is for storing program, and the processor 140 executes described program after receiving and executing instruction, aftermentioned
Method performed by the terminal device 100 that the stream process that any embodiment of the embodiment of the present invention discloses defines can be applied to handle
In device 140, or realized by processor 140.
Processor 140 may be a kind of IC chip, the processing capacity with signal.Above-mentioned processor 140 can
To be general processor, including central processing unit (Central Processing Unit, abbreviation CPU), network processing unit
(Network Processor, abbreviation NP) etc.；Can also be digital signal processor (DSP), specific integrated circuit (ASIC),
Field programmable gate array (FPGA) either other programmable logic device, discrete gate or transistor logic, discrete hard
Part component.It may be implemented or execute disclosed each method, step and the logic diagram in the embodiment of the present invention.General processor
It can be microprocessor or the processor 140 be also possible to any conventional processor 140 etc..
Various input/output devices are couple processor 140 and memory 120 by the Peripheral Interface 150.Some
In embodiment, Peripheral Interface 150, processor 140 and storage control 130 can be realized in one single chip.Other one
In a little examples, they can be realized by independent chip respectively.
Input-output unit 160 is used to be supplied to the friendship that user input data realizes user and the data collection terminal 100
Mutually.The input-output unit 160 may be, but not limited to, mouse and keyboard etc..
Audio unit 170 provides a user audio interface, may include one or more microphones, one or more raises
Sound device and voicefrequency circuit.
Display unit 180 provides an interactive interface (such as user's operation circle between data collection terminal 100 and user
Face) or for display image data give user reference.In the present embodiment, the display unit 180 can be liquid crystal display
Or touch control display.It can be the capacitance type touch control screen or resistance of support single-point and multi-point touch operation if touch control display
Formula touch screen etc..Single-point and multi-point touch operation is supported to refer to that touch control display can sense on the touch control display one
Or at multiple positions simultaneously generate touch control operation, and the touch control operation that this is sensed transfer to processor 140 carry out calculate and
Processing.
The communication unit 190 is used for by establishing connection between the network and multiple cameras, thus described in realizing
Communication connection between terminal device 100 and multiple cameras, and for passing through the network sending and receiving data.
It referring to figure 2., is the process signal of the recognition methods of chinese character in a kind of license plate provided in an embodiment of the present invention
Figure, this method comprises:
Step S110 receives license plate image.
The terminal device 100 is connect with the multiple cameras being installed on road, and multiple camera acquires vehicular traffic
License plate image after be sent to terminal device 100.
Step S120 calculates the confidence level of chinese character in the license plate image.
Specifically, since license plate number is usually the character string of letter, number and Chinese character composition, terminal device
100, which are split processing to received license plate image first, obtains single character picture to get to single numerical character figure
Picture, alphabetic character image and chinese character image.And then terminal device 100 will use the method for template matching one by one to numeric word
Symbol image, alphabetic character image and chinese character image are identified, corresponding numerical character, alphabetic character and Chinese character are obtained
Character is matched the single character picture that segmentation obtains with template one by one, if matching, the corresponding character of template is to know
Other result.Its template uses the template that the ratio of width to height is 25*50 for 0.5, size according to national standard, Chinese character, letter and
Number is respectively according to national standard file design at three class templates.
After identifying to character each in license plate image, confidence level need to be calculated to the recognition result of each character, this is set
Reliability is used to characterize the accuracy of recognition result, and confidence level is higher, then shows that recognition result is more accurate.It should be noted that
Letter, Chinese character are consistent with the algorithm of the confidence level of number, this certainty factor algebra for sentencing Chinese character is illustrated.The Chinese character
Certainty factor algebra are as follows:
Wherein, Mod (m, n) is template data, i.e. Chinese character template, template size 25*50, and template passes through two-value
Change processing, foreground pixel value are 1,1 for characterizing white, and background pixel value 0,0 is for characterizing black；The Img (m, n) is
Divide obtained chinese character, which, which carried out scaling processing, makes it keep same size with Chinese character template, i.e., at this time
The size of the chinese character is 25*50, which also passes through binary conversion treatment, i.e. the prospect of the chinese character is 1, background
It is 0；And then in the case where keeping chinese character and Chinese character template size and image format unanimous circumstances, chinese character is calculated
Confidence level is Char_Confidence, which characterizes the confidence level of individual Chinese character character, works as chinese character
When exactly matching with template, the value of Char_Confidence is 1250, it is readily appreciated that, the value of the Char_Confidence is got over
Greatly, then the matching degree of chinese character and template is higher, and recognition result is more credible.In the embodiment of the present invention, pass through template matching
Method license plate is identified, in addition to this it is possible to be identified by deep learning network to license plate.
The chinese character is arranged if the confidence level of the chinese character is lower than preset matching threshold value by step S130
For goal-selling character, the preset matching threshold value and goal-selling character are that real-time calculate updates.
Chinese character in the license plate typically represents specific province, as characterized Chongqing when chinese character is Chongqing in license plate,
Since vehicle has biggish circulation, then for the vehicle of certain area, the province that the chinese character in license plate represents can
It can be other provinces province, it is also possible to the province Ben Sheng, it is readily appreciated that, a possibility that representing the province Ben Sheng, is higher.The preset matching
Threshold value is that the chinese character in license plate characterizes the skewed popularity in the province Ben Sheng, if calculating the confidence level of chinese character lower than default
With threshold value, then show that the chinese character may identify mistake, in view of larger a possibility that occurring this province license plate in the region Ben Sheng,
This then may be identified that the chinese character of mistake is set as goal-selling character, which is the province Ben Sheng, is such as changed.
On the contrary, no matter showing that chinese character identification is correct if the confidence level for calculating chinese character is greater than preset matching threshold value
Chinese character identifies as the province Ben Sheng or other provinces province, reservation recognition result, so by Letter identification result in license plate,
Digital recognition result and chinese character recognition result arrange to obtain final license plate recognition result according to predetermined order.
It can be seen that analyzing the confidence level of the chinese character to license plate in this programme, set if confidence level is lower than
Fixed preset matching threshold value, then set the province Ben Sheng for the chinese character, to reduce false recognition rate, improves license plate and integrally identifies
Accuracy.In addition, in the present invention, the preset matching threshold value and goal-selling character are all real-time updates, with further
Improve the accuracy of judgement.
Referring to figure 3., be a kind of goal-selling character provided in an embodiment of the present invention calculation method flow diagram,
This method comprises:
Step S210 calculates the confidence level of the license plate image.
Since the license plate image is made of multiple characters, then the confidence level of entire license plate is the addition of single character, the i.e. vehicle
The calculation of the confidence level of board image are as follows:
Wherein, Char_Confidence is the confidence level of single character, and Plate_Confidence is entire license plate image
Confidence level.
Step S220, if the confidence level of the license plate image is greater than the chinese character in first threshold and the license plate image
Confidence level be greater than second threshold, then choosing chinese character in the license plate image is target characters character.
The first threshold be Plate_Thres, the Plate_Thres be characterize the license plate image recognition result be credible
Effective threshold value, the second threshold are Chn_Thres, which is the knowledge for characterizing the chinese character in the license plate image
Other result is credible effective threshold value.If meet Plate_Confidence >=Plate_Thres and Char_Confidence >
The confidence level that Chn_Thres, the i.e. confidence level of license plate image are greater than first threshold and the chinese character in license plate image is greater than the
Two threshold values then show that the recognition result of each chinese character in the license plate image is credible, and the recognition result of entire license plate image
Also to be credible, then the chinese character for choosing the license plate image is target characters character.It is readily comprehensible, since license plate may be out
In same place, it is also possible to may include multiple identical Chinese Characters in the target characters character therefore for different places
Symbol, it is also possible to include multiple and different chinese characters.
Step S230 chooses the target characters character of the first predetermined quantity from the chinese character of multiple license plate images.
Make i other words choosing a certain number of target characters characters in multiple target characters characters that step S220 is determined
For the data of subsequent statistical.
Step S240, choosing the chinese character that frequency of occurrence is most in the target characters character is goal-selling character.
Specifically, filtering out the most Chinese Character of frequency of occurrence in the target characters character of the first predetermined quantity of selection
Symbol is goal-selling character.
It should be noted that when the method for determination of the goal-selling character is that terminal device 100 just brings into operation, because eventually
When end equipment 100 just brings into operation, without more license plate image data, therefore, by a certain number of license plate images
It is the province Ben Sheng that the most chinese character of frequency of occurrence is chosen in chinese character, but increasingly with the license plate image of acquisition
More, data volume is increasing, then needs the data to the determining goal-selling character to be updated, to correct goal-selling word in real time
Symbol improves accuracy rate.
Step S250, interval predetermined time increase the target characters character of the second predetermined quantity newly.
Specifically, increasing the target characters character of the second predetermined quantity at interval of a period of time, which can
To need to be configured according to user, for example 100 or 200.
Step S260, from the target characters character of increased second predetermined quantity and the target characters word of the first predetermined quantity
It is goal-selling character that the most chinese character of frequency of occurrence is chosen in the summation of symbol.
Goal-selling character before is the Chinese that frequency of occurrence is most in the target characters character for choose the first predetermined quantity
Word character is as goal-selling character, in order to carry out real-time update to goal-selling character, then in the mesh of the first predetermined quantity
The most Chinese of frequency of occurrence is chosen based on the sample for the target characters character for increasing the second predetermined quantity on mark chinese character
Word character is goal-selling character.The target characters character of the first predetermined quantity is 500 as before, the second newly-increased predetermined quantity
Target characters character be 100, the target characters character amount according to 500 determines goal-selling character for the first time, second according to
Goal-selling character is determined according to 600 target characters character amount, and target characters character amount of the third time then according to 700 is determining default
Target character, and so on.Since the sample size is sequentially increased, and is all to increase on the basis of the original in real time, then preferably
Real-time has been reacted, ensure that the accuracy that goal-selling character determines.
Referring to figure 4., be a kind of preset matching threshold value provided in an embodiment of the present invention calculation method flow diagram,
This method comprises:
Step S310, the confidence level for choosing chinese character in the license plate image are greater than the chinese character of second threshold for system
It counts.
The second threshold is Chn_Thres, which is the identification knot for characterizing the chinese character in the license plate image
Fruit is credible effective threshold value, and the confidence level for further choosing chinese character in a certain number of license plate images is greater than second threshold
Chinese character be statistical data.
Step S320, by the confidence level of each of statistical data chinese character successively with initial matching threshold value into
Row compares, and sets goal-selling character for the chinese character that the confidence level of the chinese character is less than initial matching threshold value.
Specifically, the initial matching threshold value is a threshold value of the customized setting of user, for characterizing chinese character characterization
The skewed popularity in the province Ben Sheng shows that the chinese character may be known if the confidence level of chinese character is less than the initial matching threshold value
Not mistake may then be mistakenly identified as other provinces province, and then set goal-selling character for the chinese character, the goal-selling word
The province province Fu Biaozhengben.
Step S330, counting chinese character identical with goal-selling character in the statistical data is the first data.
Since the goal-selling character characterizes the province Ben Sheng, then count in the statistical data as the Chinese character in the characterization province Ben Sheng
The quantity of character is the first data, which is indicated with Num_Local.
Step S340, count in the statistical data with the different chinese character of goal-selling character be the second data.
Since the goal-selling character characterizes the province Ben Sheng, then the Chinese Character that other provinces province is characterized in the statistical data is counted
The quantity of symbol is the second data, which is indicated with Num_Others.
Step S350, the initial matching threshold value according to the proportion adjustment of first data and the second data are preset
Matching threshold.
Specifically, the ratio of first data and the second data is calculated, if Num_Others is 0 or Num_Local/
Num_Others >=10 then show that there are many chinese character for characterizing the province Ben Sheng, then can tune up to obtain by initial matching threshold value pre-
If matching threshold, if Num_Local/Num_Others>=5 and Num_Local/Num_Others<10, show to characterize this province
The chinese character in province is more, then can slightly increase initial matching threshold value and obtain preset threshold, in other words, tunes up initial
Degree with threshold value is determined according to the first data and the range of the second data ratio；If Num_Local/Num_Others < 5,
Show that the chinese character for characterizing the province Ben Sheng is less, then turns initial matching threshold value down to obtain preset matching threshold value.
Since the preset matching threshold value is to adjust to obtain to initial matching threshold value according to the data currently acquired, with the later period
Data volume increases, and the data that need to supplement update recalculate preset matching threshold value as basic data, to preset matching threshold value
It is updated.
The update method of the preset matching threshold value as a result, are as follows:
If the statistical data for calculating preset matching threshold value is 500 license plate images, setting is spaced from the statistical data
The corresponding chinese character of a license plate image is randomly choosed in the license plate image of quantity to replace forward one in 500 license plate images
The corresponding chinese character of license plate image is opened, with the statistical data updated, and then updated statistical data is analyzed
To be adjusted to preset matching threshold value.It such as chooses any one in 5 license plate images after statistical data 500 opens license plate image
, due to having determined that the corresponding chinese character in the province Ben Sheng before, then it can easily tell the license plate figure of selection
As the upper chinese character characterization province Ben Sheng or other provinces province, and then 500 systems are replaced with the license plate image selected
A forward license plate image, obtains updated statistical data in counting.
Further updated statistical data is analyzed, i.e., respectively in the statistical data after statistical updating with default mesh
It is third data that marking-up, which accords with identical chinese character, and third data be to characterize the quantity of the chinese character in the province Ben Sheng, and is preset
The different chinese character of target character is the 4th data, and the 4th data are to characterize the quantity of the chinese character in other provinces province, in turn
Preset matching threshold value is adjusted again according to the ratio of the third data and the 4th data, to realize to preset matching threshold value
Update.It should be noted that the setting quantity prevailing value is smaller, to reach in real time to the update of preset matching threshold value
Effect.
It can be seen that chinese character recognition methods passes through to default as reference frame in the license plate that this programme provides
Real-time update is carried out with threshold value and goal-selling character, to improve the accuracy of Car license recognition.It should be noted that when default
Target character characterizes the chinese character in the province Ben Sheng, makes after using for a period of time, will be to acquired multiple license plate images
In chinese character emptied, using again according to license plate image choose chinese character as basis data, according to identical side
Method redefines goal-selling character and preset matching threshold value, to realize the continuous updating of entire data.When recalculating, if
The chinese character in the determining characterization province Ben Sheng is as before, predetermined preset matching threshold value may be used, if not
Together, then preset matching threshold value is redefined on the basis of initial matching threshold value.
It referring to figure 5., is the function mould of the identification device 110 of chinese character in a kind of license plate provided in an embodiment of the present invention
Block schematic diagram, the device include receiving module 111, computing module 112, choose module 113, statistical module 114, setup module
115。
Receiving module 111, for receiving license plate image.
In embodiments of the present invention, step S110 can be executed by receiving module 111.
Computing module 112, for calculating the confidence level of chinese character in the license plate image.
In embodiments of the present invention, step S120 can be executed by computing module 112.
Module 113 is chosen, for choosing the target of the first predetermined quantity from the chinese character of multiple license plate images
Chinese character；Choosing the chinese character that frequency of occurrence is most in the target characters character is goal-selling character, described default
Matching threshold and goal-selling character are that real-time calculate updates.
In embodiments of the present invention, step S210~S260 can be executed by selection module 113.
Statistical module 114, the confidence level for choosing chinese character in the license plate image are greater than the Chinese character of second threshold
Character is statistical data；The confidence level of each of statistical data chinese character is successively carried out with initial matching threshold value
Compare, sets goal-selling character for the chinese character that the confidence level of the chinese character is less than initial matching threshold value；Statistics
Chinese character identical with goal-selling character is the first data in the statistical data；Count in the statistical data and default
The different chinese character of target character is the second data；According to the proportion adjustment of first data and the second data just
Beginning matching threshold obtains preset matching threshold value.
In embodiments of the present invention, step S310~S350 can be executed by statistical module 114.
Setup module 115, if the confidence level for the chinese character is lower than preset matching threshold value, by man's word
Symbol is set as goal-selling character.
In embodiments of the present invention, step S130 can be executed by setup module 115.
Since the recognition methods part of chinese character has been described in license plate, details are not described herein.
In conclusion recognition methods, device and the terminal of chinese character are set in a kind of license plate provided in an embodiment of the present invention
Standby, the recognition methods of chinese character and device are applied to terminal device in the license plate, the recognition methods of chinese character in the license plate
Include: reception license plate image, and then calculate the confidence level of each chinese character in the license plate image, if the chinese character is set
Reliability is lower than preset matching threshold value, then sets goal-selling character for the chinese character.Wherein, if the confidence of the chinese character
Degree is lower than preset matching threshold value, shows the recognition result inaccuracy of the chinese character, therefore, sets default for the chinese character
Target character, to improve the accuracy rate of chinese character identification.It should be noted that the preset matching threshold value and goal-selling word
Symbol is all real-time update, therefore, by that using the data as standard, can greatly improve the accurate of chinese character in license plate
Rate.
In several embodiments provided herein, it should be understood that disclosed device and method can also pass through
Other modes are realized.The apparatus embodiments described above are merely exemplary, for example, flow chart and block diagram in attached drawing
Show the device of multiple embodiments according to the present invention, the architectural framework in the cards of method and computer program product,
Function and operation.In this regard, each box in flowchart or block diagram can represent the one of a module, section or code
Part, a part of the module, section or code, which includes that one or more is for implementing the specified logical function, to be held
Row instruction.It should also be noted that function marked in the box can also be to be different from some implementations as replacement
The sequence marked in attached drawing occurs.For example, two continuous boxes can actually be basically executed in parallel, they are sometimes
It can execute in the opposite order, this depends on the function involved.It is also noted that every in block diagram and or flow chart
The combination of box in a box and block diagram and or flow chart can use the dedicated base for executing defined function or movement
It realizes, or can realize using a combination of dedicated hardware and computer instructions in the system of hardware.
In addition, each functional module in each embodiment of the present invention can integrate one independent portion of formation together
Point, it is also possible to modules individualism, an independent part can also be integrated to form with two or more modules.
It, can be with if the function is realized and when sold or used as an independent product in the form of software function module
It is stored in a computer readable storage medium.Based on this understanding, technical solution of the present invention is substantially in other words
The part of the part that contributes to existing technology or the technical solution can be embodied in the form of software products, the meter
Calculation machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be a
People's computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.
And storage medium above-mentioned includes: that USB flash disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), arbitrary access are deposited
The various media that can store program code such as reservoir (RAM, Random Access Memory), magnetic or disk.It needs
Illustrate, herein, relational terms such as first and second and the like be used merely to by an entity or operation with
Another entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this realities
The relationship or sequence on border.Moreover, the terms "include", "comprise" or its any other variant are intended to the packet of nonexcludability
Contain, so that the process, method, article or equipment for including a series of elements not only includes those elements, but also including
Other elements that are not explicitly listed, or further include for elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, article or equipment in there is also other identical elements.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.It should also be noted that similar label and letter exist
Similar terms are indicated in following attached drawing, therefore, once being defined in a certain Xiang Yi attached drawing, are then not required in subsequent attached drawing
It is further defined and explained.
The above description is merely a specific embodiment, but scope of protection of the present invention is not limited thereto, any
Those familiar with the art in the technical scope disclosed by the present invention, can easily think of the change or the replacement, and should all contain
Lid is within protection scope of the present invention.Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (10)
1. the recognition methods of chinese character in a kind of license plate, which is characterized in that the described method includes:
Receive license plate image；
Calculate the confidence level of chinese character in the license plate image；
If the confidence level of the chinese character is lower than preset matching threshold value, goal-selling word is set by the chinese character
Symbol, the preset matching threshold value and goal-selling character are that real-time calculate updates.
2. the method as described in claim 1, which is characterized in that the calculation of the goal-selling character are as follows:
The target characters character of the first predetermined quantity is chosen from the chinese character of multiple license plate images；
Choosing the chinese character that frequency of occurrence is most in the target characters character is goal-selling character.
3. method according to claim 2, which is characterized in that the update method of the goal-selling character includes:
It is spaced the target characters character of predetermined time newly-increased second predetermined quantity；
It is selected from the summation of the target characters character of the target characters character and the first predetermined quantity of increased second predetermined quantity
The chinese character for taking frequency of occurrence most is goal-selling character.
4. method as claimed in claim 2 or claim 3, which is characterized in that the step of selection target characters character includes:
Calculate the confidence level of the license plate image；
If the confidence level of the license plate image is greater than first threshold and the confidence level of the chinese character in the license plate image is greater than
Second threshold, then choosing the chinese character in the license plate image is target characters character.
5. the method as described in claim 1, which is characterized in that the calculation of the preset matching threshold value are as follows:
It is statistical data that the confidence level for choosing chinese character in the license plate image, which is greater than the chinese character of second threshold,；
The confidence level of each of statistical data chinese character is successively compared with initial matching threshold value, it will be described
The chinese character that the confidence level of chinese character is less than initial matching threshold value is set as goal-selling character；
Counting chinese character identical with goal-selling character in the statistical data is the first data；
Count in the statistical data with the different chinese character of goal-selling character be the second data；
The initial matching threshold value according to the proportion adjustment of first data and the second data obtains preset matching threshold value.
6. method as claimed in claim 5, which is characterized in that the update method of the preset matching threshold value are as follows:
The corresponding chinese character of one vehicle image of random selection replaces the statistics from the vehicle image of interval setting quantity
The corresponding chinese character of forward license plate image in data, obtains updated statistical data；
Calculating chinese character identical with goal-selling character in updated statistical data is third data；
Calculating chinese character different from goal-selling character in updated statistical data is the 4th data；
The preset matching threshold value is updated according to the ratio of the third data and the 4th data.
7. the identification device of chinese character in a kind of license plate, which is characterized in that described device includes:
Receiving module, for receiving license plate image；
Computing module calculates the confidence level of chinese character in the license plate image；
The chinese character is arranged if the confidence level for the chinese character is lower than preset matching threshold value for setup module
For goal-selling character, the preset matching threshold value and goal-selling character are that real-time calculate updates.
8. device as claimed in claim 7, which is characterized in that described device further includes choosing module, and the selection module is also
For:
The target characters character of the first predetermined quantity is chosen from the chinese character of multiple license plate images；
Choosing the chinese character that frequency of occurrence is most in the target characters character is goal-selling character.
9. device as claimed in claim 7, which is characterized in that described device further includes statistical module, and the statistical module is also
For:
It is statistical data that the confidence level for choosing chinese character in the license plate image, which is greater than the chinese character of second threshold,；
The confidence level of each of statistical data chinese character is successively compared with initial matching threshold value, it will be described
The chinese character that the confidence level of chinese character is less than initial matching threshold value is set as goal-selling character；
Counting chinese character identical with goal-selling character in the statistical data is the first data；
Count in the statistical data with the different chinese character of goal-selling character be the second data；
The initial matching threshold value according to the proportion adjustment of first data and the second data obtains preset matching threshold value.
10. a kind of terminal device, which is characterized in that the terminal device includes memory, processor, and the memory is for depositing
Computer program code is stored up, the processor is for executing the computer program code being stored in the memory to realize such as
The recognition methods of chinese character in license plate described in claim 1-6 any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810071711.6A CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810071711.6A CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110084232A true CN110084232A (en) | 2019-08-02 |
| CN110084232B CN110084232B (en) | 2021-01-29 |
Family
ID=67411908
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810071711.6A Active CN110084232B (en) | 2018-01-25 | 2018-01-25 | Method and device for recognizing Chinese characters in license plate and terminal equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110084232B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111639640A (en) * | 2020-04-24 | 2020-09-08 | 深圳市金溢科技股份有限公司 | License plate recognition method, device and equipment based on artificial intelligence |
| CN113496227A (en) * | 2020-04-08 | 2021-10-12 | 顺丰科技有限公司 | Training method and device of character recognition model, server and storage medium |
| CN113657372A (en) * | 2020-05-12 | 2021-11-16 | 中移(上海)信息通信科技有限公司 | License plate recognition method, device, equipment and computer storage medium |
| CN113723422A (en) * | 2021-09-08 | 2021-11-30 | 重庆紫光华山智安科技有限公司 | License plate information determination method, system, device and medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102375991A (en) * | 2010-08-24 | 2012-03-14 | 北京中星微电子有限公司 | Classifier training method and device, and character recognition method and device |
| CN103390156A (en) * | 2012-11-05 | 2013-11-13 | 深圳市捷顺科技实业股份有限公司 | License plate recognition method and device |
| CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
| CN104065920A (en) * | 2014-06-10 | 2014-09-24 | 北京中芯丙午媒体科技有限公司 | Vehicle monitoring and tracking method, system and server |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| 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 |
| CN105512660A (en) * | 2015-11-27 | 2016-04-20 | 浙江宇视科技有限公司 | License number identification method and device |
| CN106097722A (en) * | 2016-06-16 | 2016-11-09 | 广州地理研究所 | Video is utilized to carry out the system and method for trackside parking stall automatization supervision |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| US20170300786A1 (en) * | 2015-10-01 | 2017-10-19 | Intelli-Vision | Methods and systems for accurately recognizing vehicle license plates |
| CN107273896A (en) * | 2017-06-15 | 2017-10-20 | 浙江南自智能科技股份有限公司 | A kind of car plate detection recognition methods based on image recognition |
-
2018
- 2018-01-25 CN CN201810071711.6A patent/CN110084232B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102375991A (en) * | 2010-08-24 | 2012-03-14 | 北京中星微电子有限公司 | Classifier training method and device, and character recognition method and device |
| CN103390156A (en) * | 2012-11-05 | 2013-11-13 | 深圳市捷顺科技实业股份有限公司 | License plate recognition method and device |
| CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
| CN104065920A (en) * | 2014-06-10 | 2014-09-24 | 北京中芯丙午媒体科技有限公司 | Vehicle monitoring and tracking method, system and server |
| CN104464302A (en) * | 2014-12-22 | 2015-03-25 | 南京中兴力维软件有限公司 | License plate recognition and intelligent error correction method and system |
| 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 |
| US20170300786A1 (en) * | 2015-10-01 | 2017-10-19 | Intelli-Vision | Methods and systems for accurately recognizing vehicle license plates |
| CN105512660A (en) * | 2015-11-27 | 2016-04-20 | 浙江宇视科技有限公司 | License number identification method and device |
| CN106097722A (en) * | 2016-06-16 | 2016-11-09 | 广州地理研究所 | Video is utilized to carry out the system and method for trackside parking stall automatization supervision |
| CN106845478A (en) * | 2016-12-30 | 2017-06-13 | 同观科技（深圳）有限公司 | The secondary licence plate recognition method and device of a kind of character confidence level |
| CN107273896A (en) * | 2017-06-15 | 2017-10-20 | 浙江南自智能科技股份有限公司 | A kind of car plate detection recognition methods based on image recognition |
Non-Patent Citations (1)
| Title |
|---|
| 吴炜等: "一种基于模糊模板匹配的车牌汉字识别方法", 《信息技术与网络安全》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113496227A (en) * | 2020-04-08 | 2021-10-12 | 顺丰科技有限公司 | Training method and device of character recognition model, server and storage medium |
| CN111639640A (en) * | 2020-04-24 | 2020-09-08 | 深圳市金溢科技股份有限公司 | License plate recognition method, device and equipment based on artificial intelligence |
| CN111639640B (en) * | 2020-04-24 | 2023-11-14 | 深圳市金溢科技股份有限公司 | License plate recognition method, device and equipment based on artificial intelligence |
| CN113657372A (en) * | 2020-05-12 | 2021-11-16 | 中移(上海)信息通信科技有限公司 | License plate recognition method, device, equipment and computer storage medium |
| CN113657372B (en) * | 2020-05-12 | 2024-05-07 | 中移(上海)信息通信科技有限公司 | License plate recognition method, device, equipment and computer storage medium |
| CN113723422A (en) * | 2021-09-08 | 2021-11-30 | 重庆紫光华山智安科技有限公司 | License plate information determination method, system, device and medium |
| CN113723422B (en) * | 2021-09-08 | 2023-10-17 | 重庆紫光华山智安科技有限公司 | License plate information determining method, system, equipment and medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110084232B (en) | 2021-01-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10896349B2 (en) | Text detection method and apparatus, and storage medium | |
| EP3806064B1 (en) | Method and apparatus for detecting parking space usage condition, electronic device, and storage medium | |
| CN110084232A (en) | The recognition methods of chinese character, device and terminal device in license plate | |
| CN108921206B (en) | Image classification method and device, electronic equipment and storage medium | |
| CN109214002A (en) | A kind of transcription comparison method, device and its computer storage medium | |
| CN108932510A (en) | A kind of rubbish detection method and device | |
| CN108985169B (en) | Shop cross-door operation detection method based on deep learning target detection and dynamic background modeling | |
| KR20210110823A (en) | Image recognition method, training method of recognition model, and related devices and devices | |
| CN107682348A (en) | DGA domain name Quick method and devices based on machine learning | |
| CN106295498B (en) | Remote sensing image target area detection device and method | |
| CN112017171B (en) | Image processing index evaluation method, system, equipment and medium | |
| CN108390883A (en) | Recognition methods, device and the terminal device of brush popularity user | |
| CN109257390A (en) | Detection method, device and the electronic equipment of CC attack | |
| CN109034048A (en) | Face recognition algorithms models switching method and apparatus | |
| CN108986125A (en) | Object edge extracting method, device and electronic equipment | |
| CN109241962A (en) | A kind of character identifying method and device | |
| US9286217B2 (en) | Systems and methods for memory utilization for object detection | |
| CN111950345A (en) | Camera identification method and device, electronic equipment and storage medium | |
| CN110796039A (en) | Face flaw detection method and device, electronic equipment and storage medium | |
| CN108831158A (en) | It disobeys and stops monitoring method, device and electric terminal | |
| CN109492573A (en) | A kind of pointer read method and device | |
| CN112364918A (en) | Abnormality recognition method, terminal, and computer-readable storage medium | |
| CN115424193A (en) | Training image information processing method and system | |
| CN109767457A (en) | Online multi-instance learning method for tracking target, terminal device and storage medium | |
| CN111639640B (en) | License plate recognition method, device and equipment based on artificial intelligence |
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 |