CN108205670A - A kind of licence plate recognition method and device - Google Patents
A kind of licence plate recognition method and device Download PDFInfo
- Publication number
- CN108205670A CN108205670A CN201611170215.3A CN201611170215A CN108205670A CN 108205670 A CN108205670 A CN 108205670A CN 201611170215 A CN201611170215 A CN 201611170215A CN 108205670 A CN108205670 A CN 108205670A
- Authority
- CN
- China
- Prior art keywords
- character
- zone
- identified
- region
- identification result
- 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
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Character Discrimination (AREA)
- Character Input (AREA)
Abstract
The embodiment of the present application provides a kind of licence plate recognition method and device.The method includes:First, the license plate image region of the number-plate number to be identified is obtained, character recognition is carried out to the license plate image region, obtains the first character identification result;Then, it has identified successful character according in first character identification result, has determined the character zone to be identified for needing to be recognized in the license plate image region, and character recognition is carried out to the character zone to be identified, obtain the second character identification result;Finally according to first character identification result and the second character identification result, the corresponding number-plate number in the license plate image region is obtained.Car license recognition is carried out using scheme provided by the embodiments of the present application, the efficiency of Car license recognition process can be improved.
Description
Technical field
This application involves technical field of intelligent traffic, more particularly to a kind of licence plate recognition method and device.
Background technology
Car plate is vehicle " identity card ", is an important information for being different from other motor vehicles.License plate recognition technology
It has been widely used in the scenes such as bayonet, parking lot and electronic police, to obtain the number plate information of vehicle in scene, in public security
Numerous aspects such as management play the power of " intelligent transportation algorithm ".
License plate area is usually made of foreground part (character portion) and background parts (background color).Due to from practical application field
When license plate image region is obtained in scape, for car plate background parts there may be pattern, which may be since there are mud on car plate
Caused by point etc. is stained, it is also possible to be due to caused by watermark existing for car plate background parts.For example, part shown in FIG. 1
In car plate example, car plate background parts are there are irregular watermark, and the shape of watermark, color and position have diversity.
The presence of pattern will be influenced to the number-plate number identification process of this kind of license plate image in license plate image region.
In the prior art, identify the above-mentioned type car plate the number-plate number when, generally directed to need identify the number-plate number
License plate image region, it is matched to the multiple plate templates pre-saved one by one, and then identify the number-plate number, wherein, this
A little plate templates are built according to the car plate of the above-mentioned type.Detailed process is:According to selected plate template, to license plate image
Character in region carries out Character segmentation, and carries out character recognition to each character zone after segmentation.If character recognition into
Work(, then it is assumed that character identification result is determined as the corresponding license plate number in license plate image region by above-mentioned plate template successful match
Code.If character recognition is unsuccessful, another plate template is selected, is repeated the above process.
Since the type with figuratum car plate is more, in order to can recognize that the car plate of these types, it usually needs be directed to
Each type of car plate builds plate template, thus needs to build a large amount of plate templates.
Under normal conditions, it when carrying out Car license recognition using the above method, can recognize that with figuratum license plate image area
The number-plate number in domain.But due to needing to match a large amount of plate templates, and matching process will completely perform a word every time
Symbol segmentation and character recognition process, Car license recognition process efficiency are relatively low.
Invention content
The embodiment of the present application has been designed to provide a kind of licence plate recognition method and device, to improve Car license recognition process
Efficiency.Specific technical solution is as follows.
In order to achieve the above object, this application discloses a kind of licence plate recognition method, the method includes:
Obtain the license plate image region of the number-plate number to be identified;
Character recognition is carried out to the license plate image region, obtains the first character identification result;
Successful character has been identified according in first character identification result, determines to need in the license plate image region
The character zone to be identified being recognized;
Character recognition is carried out to the character zone to be identified, obtains the second character identification result;
According to first character identification result and the second character identification result, it is corresponding to obtain the license plate image region
The number-plate number.
Optionally, it is described to have identified successful character according in first character identification result, determine the car plate figure
As needing the character zone to be identified being recognized in region, including:
It determines to have identified corresponding first character zone of successful character in first character identification result, determine described
Corresponding second character zone of unidentified successful character in first character identification result;
The second character zone of preset condition will be met, be determined as needing to be recognized in the license plate image region
Character zone to be identified;
Wherein, the preset condition includes at least one of situations below:
The distance between first character zone of target character region both sides first is more than predetermined threshold value, the target character
Region is:One in second character zone;
The size of the target character region is more than size threshold, and the size threshold is according to first character zone
The threshold value that determines of average-size.
Optionally, it is described that character recognition is carried out to the character zone to be identified, obtain the second character identification result, packet
It includes:
Character segmentation is carried out to the character zone to be identified, obtains target character region;
Character recognition is carried out to the target character region, obtains the second character identification result.
Optionally, it is described that Character segmentation is carried out to the character zone to be identified, target character region is obtained, including:
It determines to have identified corresponding first character zone of successful character in first character identification result;
Obtain the average-size of first character zone;
According to the average-size, Character segmentation is carried out to the character zone to be identified, obtains target character region.
Optionally, it is described that character recognition is carried out to the license plate image region, the first character identification result is obtained, including:
Character segmentation is carried out to the license plate image region, obtains doubtful character zone;
Remove the non-character region in the doubtful character zone, the character zone after being screened;
Character recognition is carried out to the character zone after screening, obtains the first character identification result.
In order to achieve the above object, disclosed herein as well is a kind of license plate recognition device, described device includes:
Image-region obtains module, for obtaining the license plate image region of the number-plate number to be identified;
First character recognition module for carrying out character recognition to the license plate image region, obtains the first character recognition
As a result;
Character zone determining module for having identified successful character according in first character identification result, determines
The character zone to be identified being recognized is needed in the license plate image region;
Second character recognition module for carrying out character recognition to the character zone to be identified, obtains the second character and knows
Other result;
The number-plate number obtains module, for according to first character identification result and the second character identification result, obtaining
The corresponding number-plate number in the license plate image region.
Optionally, the character zone determining module, including:
First determination sub-module, for determining to have identified successful character corresponding in first character identification result
One character zone determines corresponding second character zone of unidentified successful character in first character identification result;
Second determination sub-module for that will meet the second character zone of preset condition, is determined as the license plate image area
The character zone to be identified being recognized is needed in domain;
Wherein, the preset condition includes at least one of situations below:
The distance between first character zone of target character region both sides first is more than predetermined threshold value, the target character
Region is:One in second character zone;
The size of the target character region is more than size threshold, and the size threshold is according to first character zone
The threshold value that determines of average-size.
Optionally, second character recognition module, including:
First segmentation submodule, for carrying out Character segmentation to the character zone to be identified, obtains target character region;
First identification submodule, for carrying out character recognition to the target character region, obtains the second character recognition knot
Fruit.
Optionally, the first segmentation submodule, including:
Determination unit, for determining to have identified corresponding first character of successful character in first character identification result
Region;
Obtaining unit, for obtaining the average-size of first character zone;
Cutting unit, for according to the average-size, carrying out Character segmentation to the character zone to be identified, obtaining mesh
Mark character zone.
Optionally, first character recognition module, including:
Second segmentation submodule, for carrying out Character segmentation to the license plate image region, obtains doubtful character zone;
Submodule is screened, for removing the non-character region in the doubtful character zone, the character area after being screened
Domain;
Second identification submodule, for carrying out character recognition to the character zone after screening, obtains the first character recognition knot
Fruit.
As seen from the above technical solution, in scheme provided by the embodiments of the present application, first to the license plate number to be identified of acquisition
The license plate image region of code carries out character recognition, the first character identification result is obtained, then according in the first character identification result
It has identified successful character, has determined the character zone to be identified for needing to be recognized in the license plate image region, and right
The character zone to be identified carries out character recognition, obtains the second character identification result.Finally, according to the first character identification result
With the second character identification result, the corresponding number-plate number in the license plate image region is obtained.
That is, the embodiment of the present application first carries out license plate image region first time character recognition, and according to for the first time
Successful character has been identified in the result of character recognition, has determined to need the character zone of secondary identification in license plate image region, and
It is identified.Therefore, the embodiment of the present application can improve Car license recognition process without matching a large amount of plate templates one by one
Efficiency.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or it will show below
There is attached drawing needed in technology description to be briefly described.It should be evident that the accompanying drawings in the following description is only this
Some embodiments of application, for those of ordinary skill in the art, without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is the part car plate example in somewhere;
Fig. 2 is a kind of flow diagram of licence plate recognition method provided by the embodiments of the present application;
Fig. 3 is the part car plate example of common type;
Fig. 4 is a kind of flow diagram of step S203 in Fig. 2;
Fig. 5 is the character zone and corresponding recognition result exemplary plot that license plate image region includes;
Fig. 6 is a kind of structure diagram of license plate recognition device provided by the embodiments of the present application.
Specific embodiment
Below in conjunction with the attached drawing in the embodiment of the present application, the technical solution in the embodiment of the present application is carried out clear, complete
Whole description.Obviously, described embodiment is only the part of the embodiment of the application, instead of all the embodiments.Base
Embodiment in the application, those of ordinary skill in the art are obtained all under the premise of creative work is not made
Other embodiment shall fall in the protection scope of this application.
The embodiment of the present application provides a kind of licence plate recognition method and device, and applied to electronic equipment, which can
To be terminal device or server etc., wherein, terminal device can include computer, tablet computer, smart mobile phone, driving recording
The equipment such as instrument.Car license recognition is carried out using the technical solution in the embodiment of the present application, the efficiency of Car license recognition process can be improved.
Below by specific embodiment, the application is described in detail.
Fig. 2 is a kind of flow diagram of licence plate recognition method provided by the embodiments of the present application, applied to electronic equipment.It should
Method includes the following steps:
Step S201:Obtain the license plate image region of the number-plate number to be identified.
Wherein, the license plate image region of the number-plate number to be identified can be understood as:Need the car plate figure of progress Car license recognition
As region.License plate image region is the image-region that car plate is included in license plate image.License plate image refers to include the car plate of vehicle
Partial image.As a kind of preferable embodiment, license plate image region can be that the outermost side frame of characters on license plate is formed
Image-region.Certainly, license plate image region can also be the region for including other image sections except characters on license plate.It is logical
Often, license plate image region can be arranged to rectangular area.
For example, the image-region shown in Fig. 1 can be as the license plate image region of the number-plate number to be identified.
Specifically, the license plate image region of the number-plate number to be identified, directly obtains or uses following
What mode obtained:The license plate image of the number-plate number to be identified is obtained, License Plate is carried out to the license plate image, obtains license plate image
Region.Wherein, the license plate image of the number-plate number to be identified can be understood as:Need the license plate image of progress Car license recognition.
The license plate image of the above-mentioned number-plate number to be identified can be the image comprising vehicle captured on road or
Image comprising vehicle of shooting etc. in parking lot.Certainly, the license plate image of the above-mentioned number-plate number to be identified, which can also be, passes through
What other modes obtained, the application is not defined the acquisition pattern of the license plate image of the number-plate number to be identified.
Electronic equipment internal as executive agent can include image capture device, can not also be set comprising Image Acquisition
It is standby.
When the electronic equipment internal as executive agent includes image capture device, electronic equipment is obtaining vehicle to be identified
During the license plate image of trade mark code, it can include:Receive the license plate image of the number-plate number to be identified of image capture device acquisition.
When the electronic equipment internal as executive agent does not include image capture device, which can be with outside
Image capture device be connected, electronic equipment can include in the license plate image for obtaining the number-plate number to be identified:Obtain image
The license plate image of the number-plate number to be identified of collecting device acquisition.
The license plate image of the number-plate number to be identified obtained can be that image capture device acquires in real time, may not be
It acquires in real time, but image capture device is collected what is stored later in advance.
Step S202:Character recognition is carried out to the license plate image region, obtains the first character identification result.
Specifically, the character in identification license plate image region, it, can be first using vertical when obtaining the first character identification result
Straight sciagraphy or connection domain method etc. are split license plate image region, obtain each character in the license plate image region
Then region is identified character zone using preset Character recognizer, obtain the first character identification result.
Wherein, above-mentioned character zone is the image-region that character may be included in license plate image region.Above-mentioned character recognition
Device can include N number of output unit, and each output unit corresponds to a character.For example, Character recognizer includes 37 output lists
Member, these output units correspond to following character respectively:10 numbers, 26 letters and 1 " unknown ".
Above-mentioned Character segmentation result generally includes multiple character zones, corresponding, and the first character identification result generally includes
Successful character and corresponding character zone, unidentified successful character and corresponding character zone are identified.
Specifically, when carrying out character recognition to each character zone, which can be inputted above-mentioned character recognition
Device, each output unit can export a confidence level, and the corresponding character of output unit that confidence level is more than predetermined threshold value is exactly
Character in the character zone.At this moment, it is believed that the character zone identifies that successfully corresponding character is just identified as the character of work(.
If the confidence level of " unknown " output unit is higher than threshold value, and the confidence level of other output units is below threshold value,
Then think that character zone identification is unsuccessful, " unknown " is unidentified successful character.
In addition, when being split to license plate image region, can car plate be obtained according to vertical projection method or connection domain method
The characteristic image of image-region pixel determines Character segmentation boundary, so as to obtain Character segmentation result from this feature image.
Therefore, apparent and consistent features are respectively provided with for all characters and car plate prospect and the preferable vehicle of background contrast's property
Board image in the license plate area for dividing this kind of license plate image, is commonly available preferable segmentation result, and then be easier to just
Really identify all characters in license plate image.
For example, Fig. 3 show a part of car plate instance graph of common type, the size base of the character of number-plate number part
This is consistent, and the spacing between adjacent character is basically identical.Wherein, the height of the size of character zone including character zone and/or
Width.As it can be seen that the character in this kind of car plate has apparent and consistent feature.In addition, car plate shown in Fig. 3 also has, there are one apparent
Foreground (i.e. the color of character portion) and the color unicity of background colour are preferable in feature, i.e. license plate area, and foreground
Comparative higher with background colour, discrimination is preferable.For example, for the car plate that number is 5 in Fig. 3, wherein, prospect is single
White, background are single black, it is seen that foreground part is respectively provided with single color with background parts in the car plate.To this kind of
When the license plate image region of car plate carries out Character segmentation, usually can preferably distinguish the boundary of each character zone, obtain compared with
Good segmentation result, and then better character identification result can be obtained.
Certainly, there is also the figuratum situation in license plate image regional background part in real life, the pattern may be by
Caused by being stained on car plate, it is also possible to which due to car plate background parts, there are caused by irregular watermark.Such as institute in Fig. 1
Showing the background parts of car plate, there are irregular watermarks.Such car plate background parts color unicity is poor, and pattern is deposited
It may lead to Characters Stuck, and then influence determining for character zone boundary, lead to not accurately divide in car plate image-region
Character zone.Therefore when carrying out character recognition to license plate image region, it is understood that there may be identification mistake.
Step S203:Successful character has been identified according in first character identification result, determines the license plate image
The character zone to be identified being recognized is needed in region.
By the description of abovementioned steps it is found that after by above-mentioned first time Car license recognition process, in license plate image region
Character may identify completely, it is also possible to do not identify completely, i.e., also there are it is unidentified go out character.Cause
This, successful character is being identified according in first character identification result, determine to need in the license plate image region into
During the character zone to be identified of the secondary identification of row, can first according to having identified successful character in the first character identification result,
Judge with the presence or absence of the character zone to be identified for needing to be recognized in license plate image region, if it is present determining institute
Character zone to be identified is stated, if it does not exist, then directly will identify that successful character is determined as in the first character identification result
The number-plate number in license plate image region.
It, can be from the first character when needing the character zone to be identified being recognized in determining license plate image region
It is determined in the corresponding character zone of unidentified successful character in recognition result.
It should be noted that when, there are during unidentified successful character, these characters correspond in the first character identification result
Character zone may not exactly need the character zone to be identified being recognized, it is also possible to rivet area in car plate,
Mud point region etc..Therefore, it is necessary to be found out from the corresponding character zone of unidentified successful character of the first character identification result
Need the character zone being recognized.
Specifically, according to successful character has been identified in the first character identification result, determine in the license plate image region
The character zone to be identified being recognized is needed, numerous embodiments can be included:
One kind is, according to the size for having identified the corresponding character zone of successful character, never to identify successful character pair
The character zone of size exception is determined in the character zone answered, the character zone of this size exception is determined as character area to be identified
Domain.
It is understood that when the background parts of license plate area there are pattern when influences when, may cause to send out between character
Raw adhesion, and then corresponding character zone is caused to become large-sized, it has been segmented in one in segmentation or character and background patterns
It rises or at least two characters is segmented in together with intermediate pattern.Therefore, character recognition is led to not by this reason
During success, the character zone that needs are recognized whether can be determined extremely according to the size of character zone.
Another kind is, according to having identified position of the corresponding character zone of successful character in license plate image region point
Cloth never identifies and character zone to be identified is determined in the corresponding character zone of successful character.
If it is understood that between having identified successful character zone, also there are character zones to be identified, then
It has identified that the spacing between successful character zone is likely to very big, spatial abnormal feature feature is presented.It therefore, can be different according to this
Normal distribution characteristics determines character zone to be identified.
Certainly, according to having identified that successful character determines that the embodiment of character zone to be identified is also very much, herein not
It enumerates.The present embodiment is not specifically limited the embodiment of the process.
Step S204:Character recognition is carried out to the character zone to be identified, obtains the second character identification result.
It should be noted that when identifying the character in character zone to be identified, the mistake identical with step S202 may be used
Journey, can also use the process different from step S202, and detailed identification process the present embodiment repeats no more.
Step S205:According to first character identification result and the second character identification result, the license plate image is obtained
The corresponding number-plate number in region.
Specifically, since there may be unidentified successful in the first character identification result and the second character identification result
Character, therefore, in order to improve the accuracy of recognition result, will usually have been identified in the first character identification result successful character and
It has identified that successful character is synthesized in second character identification result, has obtained the corresponding license plate number in the license plate image region
Code.
It more specifically, can be according to the relative position of the corresponding character zone of each character to identifying twice in synthesis
As a result identified that successful character is ranked up in, according to ranking results composite characters, the final license plate image region that obtains corresponds to
The number-plate number.
As shown in the above, the present embodiment carries out word to the license plate image region of the number-plate number to be identified of acquisition first
Symbol identification, obtains the first character identification result, then according to successful character has been identified in the first character identification result, determines institute
The character zone to be identified for needing to be recognized in license plate image region is stated, and word is carried out to the character zone to be identified
Symbol identification, obtains the second character identification result.Finally, it according to the first character identification result and the second character identification result, obtains
The corresponding number-plate number in the license plate image region.
That is, the present embodiment first carries out first time character recognition to license plate image region, and according to first time character
Successful character has been identified in the result of identification, has determined to need the character zone of secondary identification in license plate image region, and to it
It is identified.Therefore, when carrying out Car license recognition using scheme provided in this embodiment, without matching a large amount of plate templates one by one,
The efficiency of Car license recognition process can be improved.
In a kind of specific embodiment based on embodiment illustrated in fig. 2, step S203, according to first character recognition
As a result successful character has been identified in, has determined the character area to be identified for needing to be recognized in the license plate image region
Domain can carry out according to flow diagram shown in Fig. 4, include the following steps:
Step S203A:It determines to have identified corresponding first character area of successful character in first character identification result
Domain determines corresponding second character zone of unidentified successful character in first character identification result.
Step S203B:The second character zone of preset condition will be met, be determined as needing in the license plate image region into
The character zone to be identified of the secondary identification of row.
Wherein, the preset condition includes at least one of situations below:
Situation one:The distance between first character zone of target character region both sides first is more than predetermined threshold value, described
Target character region is:One in second character zone.
Predetermined threshold value can be determined previously according to the feature in a large amount of sample license plate image region.It for example, can be by sample
The mean breadth of single character zone is as predetermined threshold value in this license plate image region.
Specifically, when determining the character zone to be identified, each second character zone both sides can be first determined
One the first character zone, then calculate two the first character zones to the distance between, will distance be more than predetermined threshold value two
The second character zone between a first character zone pair, is determined as character zone to be identified.
For example, Fig. 5 show the first character identification result obtained for license plate image region, the first character identification result
In the character that has identified and it is unidentified go out character be marked below license plate image region, box table is used inside license plate image region
The corresponding character zone of each character is shown, and lists the number of each character zone with number above box.It so can be true
The number of delimiting the organizational structure is that the character zone of " 3 " is the second character zone, and can determine 3 left and right sides of character zone and the character
Two adjacent the first character zones of region 3 or so are respectively character zone 2 and character zone 4, and character zone 2 is understood through judging
The distance between character zone 4 is more than predetermined threshold value, then can determine that the character zone 3 needs are recognized
Character zone to be identified.(character of character zone that number is 5 in Fig. 5 is not shown)
Situation two:The size of the target character region is more than size threshold, and the size threshold is according to described first
The threshold value that the average-size of character zone determines.Wherein it is possible to using the average-size and the product of preset value as size threshold,
It is, of course, also possible to will be between average-size and preset value and as size threshold, this is all feasible.The present embodiment to this not
It is specifically limited.
Specifically, when determining the character zone to be identified, the size of each second character zone can be first determined, with
And the average-size of all first character zones, and using the average-size of preset value times as size threshold.Then by each
The size of two character zones is compared with the size threshold, and second character zone of the size more than the size threshold is determined
For character zone to be identified.Wherein, above-mentioned size includes at least one of width, height, the ratio of width to height.
Continue to use the example in the above situation one, character zone 1,2,4,5,6 and 7 is the first character zone, character zone 3
For the second character zone, and it can determine the mean breadth of above-mentioned first character zone.Then judge the width of the second character zone
Whether degree is more than the mean breadth of preset value times, if it is greater, then character zone 3 is determined as character zone to be identified.
In summary, the second character zone for meeting preset condition is determined as needs in license plate image region by the present embodiment
The character zone being recognized, and preset condition is related to the first character zone.And the first character zone is the first word
The corresponding character zone of successful character has been identified in symbol recognition result, therefore according to the feature of the first character zone from the second word
It accords with and character zone to be identified is determined in region, the accuracy of determination process can be improved.
Further, accuracy during character recognition is carried out to character zone to be identified in order to improve, based on shown in Fig. 2
In a kind of specific embodiment of embodiment, step S204 carries out character recognition to the character zone to be identified, obtains second
Character identification result can include following sub-step:
Sub-step 1:Character segmentation is carried out to the character zone to be identified, obtains target character region.
Specifically, the character zone to be identified can be divided into line character according to vertical projection method or connection domain method etc.
It cuts, obtains target character region.
As a kind of specific embodiment, in order to improve the accuracy of Character segmentation process, sub-step 1 is treated to described
It identifies that character zone carries out Character segmentation, obtains target character region, can include:
It determines to have identified corresponding first character zone of successful character in first character identification result, described in acquisition
The average-size of first character zone according to the average-size, carries out Character segmentation to the character zone to be identified, obtains
Target character region.
Specifically, according to the average-size, it, can be by described in when carrying out Character segmentation to the character zone to be identified
Size of the average-size as character in character zone to be identified, with reference to the corresponding upright projection characteristic value of character zone to be identified
Or connected component value, Character segmentation is carried out to the character zone to be identified.
It is understood that character zone to be identified may be character and pattern is sticked together the character zone to be formed,
It is also likely to be that character, pattern, character are sticked together the character zone to be formed jointly.If character zone to be identified can be obtained
The size of middle character, then the process that Character segmentation is carried out to the character zone to be identified just can be more accurate.
Sub-step 2:Character recognition is carried out to the target character region, obtains the second character identification result.
To sum up, in the present embodiment, first the character zone to be identified is carried out as the electronic equipment of executive agent
Then Character segmentation carries out character recognition to the character zone after segmentation, the character zone after segmentation is possible include character
Region carries out character recognition for such character zone, can improve the accuracy of character recognition process one by one.
In addition, due to including the watermark being separated from each other with character portion in the background of part license plate area, such as compiled in Fig. 1
Number for 1~3 car plate second half section comprising watermark recovery with other character adhesions, therefore, in order to improve identification process
Accuracy, the present embodiment can also include implementation below.
In a kind of specific embodiment based on embodiment illustrated in fig. 2, step S202, to the license plate image region into
Line character identifies, obtains the first character identification result, can include following sub-step:
Sub-step 1:Character segmentation is carried out to the license plate image region, obtains doubtful character zone.
Sub-step 2:Remove the non-character region in doubtful character zone, the character zone after being screened.
Wherein, non-character region can include region and/or the area of the pattern of solid color.Specifically, remove doubtful word
, can be according to the non-character identification network removal being generated in advance when according with the non-character region in region, it can also be according in region
Pixel characteristic value removal.
At the non-character region during doubtful character zone is removed according to the non-character being generated in advance identification network, can incite somebody to action
Doubtful character zone input non-character identification network obtains representing that the doubtful character zone is non-word from non-character identification network
The confidence level in region is accorded with, according to the confidence level, it is determined whether remove the doubtful character zone.It is understood that the confidence level
It is bigger, illustrate that the doubtful character zone is more likely to be non-character region.Therefore, confidence level can be deleted and be more than default confidence level
The doubtful character zone of threshold value, the character zone after being screened.Above-mentioned non-character identification network can be according to acquiring in advance
Sample character picture is trained, and above-mentioned sample character picture can include positive sample character picture and negative sample character picture,
Positive sample character picture includes non-character image.
Sub-step 3:Character recognition is carried out to the character zone after screening, obtains the first character identification result.
In summary, in the present embodiment, the is being carried out to license plate image region as the electronic equipment of executive agent
It during character recognition, can first introduce " non-character identification " process to be filtered the non-character region that segmentation obtains, reduce
The quantity of unidentified successful character zone, accuracy that is near and improving character recognition process.
Fig. 6 is a kind of structure diagram of license plate recognition device provided by the embodiments of the present application, is implemented with method shown in Fig. 2
Example is corresponding, applied to electronic equipment.Described device includes:
Image-region obtains module 601, for obtaining the license plate image region of the number-plate number to be identified;
First character recognition module 602 for carrying out character recognition to the license plate image region, obtains the first character and knows
Other result;
Character zone determining module 603, for having identified successful character according in first character identification result, really
The character zone to be identified being recognized is needed in the fixed license plate image region;
Second character recognition module 604 for carrying out character recognition to the character zone to be identified, obtains the second character
Recognition result;
The number-plate number obtains module 605, for according to first character identification result and the second character identification result, obtaining
Obtain the corresponding number-plate number in the license plate image region.
In a kind of specific embodiment based on embodiment illustrated in fig. 6, the character zone determining module 603 can be with
Including:
First determination sub-module (not shown), for determining to have identified in first character identification result successfully
Corresponding first character zone of character determines corresponding second word of unidentified successful character in first character identification result
Accord with region;
Second determination sub-module (not shown) for that will meet the second character zone of preset condition, is determined as institute
State the character zone to be identified for needing to be recognized in license plate image region;
Wherein, the preset condition includes at least one of situations below:
The distance between first character zone of target character region both sides first is more than predetermined threshold value, the target character
Region is:One in second character zone;
The size of the target character region is more than size threshold, and the size threshold is according to first character zone
The threshold value that determines of average-size.
In a kind of specific embodiment based on embodiment illustrated in fig. 6, second character recognition module 604 can be with
Including:
First segmentation submodule (not shown), for carrying out Character segmentation to the character zone to be identified, obtains
Target character region;
First identification submodule (not shown), for carrying out character recognition to the target character region, obtains the
Two character identification results.
In a kind of specific embodiment based on embodiment illustrated in fig. 6, the first segmentation submodule can include:
Determination unit (not shown), for determining to have identified successful character pair in first character identification result
The first character zone answered;
Obtaining unit (not shown), for obtaining the average-size of first character zone;
Cutting unit (not shown), for according to the average-size, word to be carried out to the character zone to be identified
Symbol segmentation, obtains target character region.
In a kind of specific embodiment based on embodiment illustrated in fig. 6, first character recognition module 602 can be with
Including:
Second segmentation submodule (not shown), for carrying out Character segmentation to the license plate image region, is doubted
Like character zone;
Submodule (not shown) is screened, for removing the non-character region in the doubtful character zone, is sieved
Character zone after choosing;
Second identification submodule (not shown), for carrying out character recognition to the character zone after screening, obtains the
One character identification result.
Since above device embodiment is obtained based on embodiment of the method, there is identical technique effect with this method,
Therefore details are not described herein for the technique effect of device embodiment.For device embodiment, since it is substantially similar to method
Embodiment, so describing fairly simple, the relevent part can refer to the partial explaination of embodiments of method.
It should be noted that herein, relational terms such as first and second and the like are used merely to a reality
Body or operation are distinguished with another entity or operation, are deposited without necessarily requiring or implying between these entities or operation
In any this practical relationship or sequence.Moreover, term " comprising ", "comprising" or any other variant be intended to it is non-
It is exclusive to include, so that process, method, article or equipment including a series of elements not only include those elements,
But also it including other elements that are not explicitly listed or further includes solid by this process, method, article or equipment
Some elements.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including
Also there are other identical elements in the process of the element, method, article or equipment.
It will appreciated by the skilled person that all or part of step in the above embodiment is can to pass through journey
Sequence instructs relevant hardware, and come what is completed, the program can be stored in computer read/write memory medium.It is designated herein
Storage medium refers to ROM/RAM, magnetic disc, CD etc..
The foregoing is merely the preferred embodiments of the application, are not intended to limit the protection domain of the application.It is all
Any modification, equivalent substitution, improvement and etc. done within spirit herein and principle are all contained in the protection domain of the application
It is interior.
Claims (10)
1. a kind of licence plate recognition method, which is characterized in that the method includes:
Obtain the license plate image region of the number-plate number to be identified;
Character recognition is carried out to the license plate image region, obtains the first character identification result;
Successful character has been identified according in first character identification result, determines to need to carry out in the license plate image region
The character zone to be identified of secondary identification;
Character recognition is carried out to the character zone to be identified, obtains the second character identification result;
According to first character identification result and the second character identification result, the corresponding car plate in the license plate image region is obtained
Number.
2. according to the method described in claim 1, it is characterized in that, described identified according in first character identification result
Successful character determines the character zone to be identified for needing to be recognized in the license plate image region, including:
It determines to have identified corresponding first character zone of successful character in first character identification result, determines described first
Corresponding second character zone of unidentified successful character in character identification result;
The second character zone of preset condition will be met, be determined as needing to be recognized in the license plate image region treats
Identify character zone;
Wherein, the preset condition includes at least one of situations below:
The distance between first character zone of target character region both sides first is more than predetermined threshold value, the target character region
For:One in second character zone;
The size of the target character region is more than size threshold, and the size threshold is according to the flat of first character zone
The threshold value that equal size determines.
3. according to the method described in claim 1, it is characterized in that, described know the character zone to be identified into line character
Not, the second character identification result is obtained, including:
Character segmentation is carried out to the character zone to be identified, obtains target character region;
Character recognition is carried out to the target character region, obtains the second character identification result.
4. according to the method described in claim 3, it is characterized in that, described divide the character zone to be identified into line character
It cuts, obtains target character region, including:
It determines to have identified corresponding first character zone of successful character in first character identification result;
Obtain the average-size of first character zone;
According to the average-size, Character segmentation is carried out to the character zone to be identified, obtains target character region.
5. according to the method described in claim 1, it is characterized in that, it is described to the license plate image region carry out character recognition,
The first character identification result is obtained, including:
Character segmentation is carried out to the license plate image region, obtains doubtful character zone;
Remove the non-character region in the doubtful character zone, the character zone after being screened;
Character recognition is carried out to the character zone after screening, obtains the first character identification result.
6. a kind of license plate recognition device, which is characterized in that described device includes:
Image-region obtains module, for obtaining the license plate image region of the number-plate number to be identified;
First character recognition module for carrying out character recognition to the license plate image region, obtains the first character identification result;
Character zone determining module for having identified successful character according in first character identification result, determines described
The character zone to be identified being recognized is needed in license plate image region;
Second character recognition module for carrying out character recognition to the character zone to be identified, obtains the second character recognition knot
Fruit;
The number-plate number obtains module, for according to first character identification result and the second character identification result, described in acquisition
The corresponding number-plate number in license plate image region.
7. device according to claim 6, which is characterized in that the character zone determining module, including:
First determination sub-module, for determining to have identified corresponding first word of successful character in first character identification result
Region is accorded with, determines corresponding second character zone of unidentified successful character in first character identification result;
Second determination sub-module for that will meet the second character zone of preset condition, is determined as in the license plate image region
Need the character zone to be identified being recognized;
Wherein, the preset condition includes at least one of situations below:
The distance between first character zone of target character region both sides first is more than predetermined threshold value, the target character region
For:One in second character zone;
The size of the target character region is more than size threshold, and the size threshold is according to the flat of first character zone
The threshold value that equal size determines.
8. device according to claim 6, which is characterized in that second character recognition module, including:
First segmentation submodule, for carrying out Character segmentation to the character zone to be identified, obtains target character region;
First identification submodule, for carrying out character recognition to the target character region, obtains the second character identification result.
9. device according to claim 8, which is characterized in that the first segmentation submodule, including:
Determination unit, for determining to have identified corresponding first character area of successful character in first character identification result
Domain;
Obtaining unit, for obtaining the average-size of first character zone;
Cutting unit, for according to the average-size, carrying out Character segmentation to the character zone to be identified, obtaining target word
Accord with region.
10. device according to claim 6, which is characterized in that first character recognition module, including:
Second segmentation submodule, for carrying out Character segmentation to the license plate image region, obtains doubtful character zone;
Submodule is screened, for removing the non-character region in the doubtful character zone, the character zone after being screened;
Second identification submodule, for carrying out character recognition to the character zone after screening, obtains the first character identification result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611170215.3A CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201611170215.3A CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108205670A true CN108205670A (en) | 2018-06-26 |
CN108205670B CN108205670B (en) | 2020-10-27 |
Family
ID=62601647
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201611170215.3A Active CN108205670B (en) | 2016-12-16 | 2016-12-16 | License plate recognition method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108205670B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109034158A (en) * | 2017-06-09 | 2018-12-18 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method, device and computer equipment |
CN111767909A (en) * | 2020-05-12 | 2020-10-13 | 合肥联宝信息技术有限公司 | Character recognition method and device and computer readable storage medium |
CN113486885A (en) * | 2021-06-17 | 2021-10-08 | 杭州鸿泉物联网技术股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2006338578A (en) * | 2005-06-06 | 2006-12-14 | Mitsubishi Electric Corp | Character recognition apparatus |
US7182492B1 (en) * | 2003-12-22 | 2007-02-27 | Robert Louis Walter | License plate system having enhanced illumination |
CN1996995A (en) * | 2006-12-29 | 2007-07-11 | 信息产业部电信传输研究所 | Control method for service sensing and its system |
CN101944174A (en) * | 2009-07-08 | 2011-01-12 | 西安电子科技大学 | Identification method of characters of licence plate |
CN103413147A (en) * | 2013-08-28 | 2013-11-27 | 庄浩洋 | Vehicle license plate recognizing method and system |
CN105335743A (en) * | 2015-10-28 | 2016-02-17 | 重庆邮电大学 | Vehicle license plate recognition method |
-
2016
- 2016-12-16 CN CN201611170215.3A patent/CN108205670B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7182492B1 (en) * | 2003-12-22 | 2007-02-27 | Robert Louis Walter | License plate system having enhanced illumination |
JP2006338578A (en) * | 2005-06-06 | 2006-12-14 | Mitsubishi Electric Corp | Character recognition apparatus |
CN1996995A (en) * | 2006-12-29 | 2007-07-11 | 信息产业部电信传输研究所 | Control method for service sensing and its system |
CN101944174A (en) * | 2009-07-08 | 2011-01-12 | 西安电子科技大学 | Identification method of characters of licence plate |
CN103413147A (en) * | 2013-08-28 | 2013-11-27 | 庄浩洋 | Vehicle license plate recognizing method and system |
CN105335743A (en) * | 2015-10-28 | 2016-02-17 | 重庆邮电大学 | Vehicle license plate recognition method |
Non-Patent Citations (2)
Title |
---|
K. MIYAMOTO: "Vehicle license-plate recognition by image analysis", 《PROCEEDINGS IECON "91: 1991 INTERNATIONAL CONFERENCE ON INDUSTRIAL ELECTRONICS, CONTROL AND INSTRUMENTATION》 * |
李坤: "车牌自动识别系统的研究与设计", 《中国优秀硕士学位论文全文数据库 信息科学辑》 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109034158A (en) * | 2017-06-09 | 2018-12-18 | 杭州海康威视数字技术股份有限公司 | A kind of licence plate recognition method, device and computer equipment |
CN111767909A (en) * | 2020-05-12 | 2020-10-13 | 合肥联宝信息技术有限公司 | Character recognition method and device and computer readable storage medium |
CN111767909B (en) * | 2020-05-12 | 2022-02-01 | 合肥联宝信息技术有限公司 | Character recognition method and device and computer readable storage medium |
CN113486885A (en) * | 2021-06-17 | 2021-10-08 | 杭州鸿泉物联网技术股份有限公司 | License plate recognition method and device, electronic equipment and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN108205670B (en) | 2020-10-27 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108073928B (en) | License plate recognition method and device | |
CN105373794B (en) | A kind of licence plate recognition method | |
CN108229466A (en) | A kind of licence plate recognition method and device | |
Xie et al. | A robust license plate detection and character recognition algorithm based on a combined feature extraction model and BPNN | |
CN106650553A (en) | License plate recognition method and system | |
CN105740886B (en) | A kind of automobile logo identification method based on machine learning | |
CN108073926B (en) | License plate recognition method and device | |
CN110163109B (en) | Lane line marking method and device | |
Paunwala et al. | A novel multiple license plate extraction technique for complex background in Indian traffic conditions | |
CN106384513A (en) | Fake-licensed car capturing system and method based on intelligent traffic | |
CN105184291B (en) | A kind of polymorphic type detection method of license plate and system | |
CN107180230B (en) | Universal license plate recognition method | |
CN108090484B (en) | License plate recognition method and device | |
Azad et al. | New method for optimization of license plate recognition system with use of edge detection and connected component | |
CN107590500A (en) | A kind of color recognizing for vehicle id method and device based on color projection classification | |
CN108205670A (en) | A kind of licence plate recognition method and device | |
CN104408431B (en) | Car money recognition methods under traffic monitoring | |
CN101369312B (en) | Method and equipment for detecting intersection in image | |
CN109389122A (en) | A kind of license plate locating method and device | |
CN108073925B (en) | License plate recognition method and device | |
CN112115800A (en) | Vehicle combination recognition system and method based on deep learning target detection | |
CN106327876A (en) | Faked plate vehicle capturing system and method based on vehicle recorder | |
CN107392115B (en) | Traffic sign identification method based on hierarchical feature extraction | |
CN106778765B (en) | License plate recognition method and device | |
Chowdhury et al. | An adaptive technique for computer vision based vehicles license plate detection system |
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 |