CN108961532A - Crown word number image processing method, device, equipment and storage medium - Google Patents
Crown word number image processing method, device, equipment and storage medium Download PDFInfo
- Publication number
- CN108961532A CN108961532A CN201710384027.9A CN201710384027A CN108961532A CN 108961532 A CN108961532 A CN 108961532A CN 201710384027 A CN201710384027 A CN 201710384027A CN 108961532 A CN108961532 A CN 108961532A
- Authority
- CN
- China
- Prior art keywords
- connected domain
- image
- noise
- predeterminable area
- crown word
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000004458 analytical method Methods 0.000 claims abstract description 45
- 238000000034 method Methods 0.000 claims abstract description 29
- 230000015654 memory Effects 0.000 claims description 18
- 238000004590 computer program Methods 0.000 claims description 3
- 238000010586 diagram Methods 0.000 description 9
- 230000006870 function Effects 0.000 description 6
- 230000011218 segmentation Effects 0.000 description 4
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000010191 image analysis Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07D—HANDLING OF COINS OR VALUABLE PAPERS, e.g. TESTING, SORTING BY DENOMINATIONS, COUNTING, DISPENSING, CHANGING OR DEPOSITING
- G07D7/00—Testing specially adapted to determine the identity or genuineness of valuable papers or for segregating those which are unacceptable, e.g. banknotes that are alien to a currency
- G07D7/20—Testing patterns thereon
- G07D7/2008—Testing patterns thereon using pre-processing, e.g. de-blurring, averaging, normalisation or rotation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
- Inspection Of Paper Currency And Valuable Securities (AREA)
Abstract
The embodiment of the invention discloses a kind of crown word number image processing method, device, equipment and storage mediums, wherein this method comprises: obtaining the bianry image in crown word number region;Connected domain analysis is carried out to foreground pixel point each in the predeterminable area of bianry image, obtains the connected domain of predeterminable area;According to the number of connected domain and the feature of image of predeterminable area, the noise connected domain in connected domain is determined;Pixel in noise connected domain is changed to background pixel point, obtains updated image.The embodiment of the present invention utilizes connected domain analysis, in conjunction with the number of connected domain in the feature of image of predeterminable area itself and the region, it can determine edge noise position, and edge noise be changed to background pixel, thereby eliminate the influence that edge noise is divided crown word number and identified.The program is handled for predeterminable area simultaneously, and it is indirect whole image is handled, time complexity is low, and treatment effeciency is high.
Description
Technical field
The present embodiments relate to image processing techniques more particularly to a kind of crown word number image processing methods, device, equipment
And storage medium.
Background technique
Existing paper money counter, cash inspecting machine, ATM machine etc. usually have cash examining function, i.e. identification paper money number, to identify paper
The coin true and false.Crown word number is identified, needs to intercept the image in crown word number region from banknote image, then therefrom extracts hat
Font size is to be identified.
But when intercepting crown word number image, other foreground points except crown word number are often truncated to, for example, extending
To the pattern in crown word number region, so before extraction sight spot when, will receive the influence of these marginal foreground points, it is accurate to lead to not
The crown word number of needs is isolated, and then influences the accurate positionin of character boundary, causes crown word number identification mistake that can not even identify,
Certain difficulty is brought for paper money recognition or false distinguishing.
Summary of the invention
The embodiment of the present invention provides a kind of crown word number image processing method, device, equipment and storage medium, can distinguish
Marginal foreground point eliminates the influence that marginal foreground point is divided crown word number and identified.
In a first aspect, the embodiment of the invention provides a kind of crown word number image processing methods, comprising:
Obtain the bianry image in crown word number region;
Connected domain analysis is carried out to foreground pixel point each in the predeterminable area of the bianry image, obtains the predeterminable area
Connected domain;
According to the number of connected domain and the feature of image of the predeterminable area, the noise connection in the connected domain is determined
Domain;
Pixel in the noise connected domain is changed to background pixel point, obtains updated image.
Further, before the bianry image for obtaining crown word number region, the method also includes: it is installed according to noise bits
The predeterminable area is set, wherein the predeterminable area includes at least partial character.
Further, it according to the number of connected domain and the feature of image of the predeterminable area, determines in the connected domain
Noise connected domain, comprising:
If the number of connected domain is one, it is determined that noise connected domain is not present;
If the number of connected domain is two and two or more, the pixel of predetermined number is extended to the predeterminable area,
Connected domain analysis is carried out to the region after extension, noise connected domain is determined according to new connected domain analysis result.
Further, noise connected domain is determined according to new connected domain analysis result, comprising:
In the case where the number of former connected domain is two, if the number of new connected domain is one, it is determined that be not present
Noise connected domain;If the number of new connected domain is two, noise connected domain is determined according to the feature of image of predeterminable area;
In the case where the number of former connected domain is three and three or more, the former connected domain is judged according to new connected domain
Intersection situation;If there is intersection, it is determined that the connected domain for not having intersection in former connected domain is noise connected domain;If do not deposited
In intersection, then noise connected domain is determined according to the feature of image of the predeterminable area.
Further, connected domain analysis is carried out to foreground pixel point each in the predeterminable area of the bianry image, comprising: adopt
Connected domain analysis is carried out to foreground pixel point each in the predeterminable area of the bianry image with eight connected region or four connected region.
Further, the bianry image in crown word number region is obtained, comprising: crown word number region is intercepted from bank note grayscale image
Grayscale image;Binaryzation is carried out to the grayscale image in the crown word number region, obtains the bianry image in the crown word number region.
Second aspect, the embodiment of the invention also provides a kind of crown word number image processing apparatus, comprising:
Image collection module, for obtaining the bianry image in crown word number region;
Connected domain analysis module carries out connected domain minute for foreground pixel point each in the predeterminable area to the bianry image
Analysis, obtains the connected domain of the predeterminable area;
Noise connected domain determining module, for according to the number of connected domain, the feature of image of the predeterminable area and default
Rule determines the noise connected domain in the connected domain;
Image update module is updated for the pixel in the noise connected domain to be changed to background pixel point
Image afterwards.
Further, the noise connected domain determining module includes:
In the case that first connected domain determination unit for the number in connected domain is one, determine that there is no noises to connect
Logical domain;
In the case that second connected domain determination unit for the number in connected domain is two and two or more, to described
Predeterminable area extends the pixel of predetermined number, connected domain analysis is carried out to the region after extension, according to new connected domain analysis knot
Fruit determines noise connected domain.
The third aspect, the embodiment of the invention also provides a kind of equipment, the equipment includes:
One or more processors;
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processing
Device realizes such as crown word number image processing method provided by any embodiment of the invention.
Fourth aspect, the embodiment of the invention also provides a kind of computer readable storage mediums, are stored thereon with computer
Program realizes such as crown word number image processing method provided by any embodiment of the invention when the program is executed by processor.
Crown word number image processing method, device, equipment and the storage medium of the embodiment of the present invention, using connected domain analysis,
In conjunction with the number of connected domain in the feature of image of predeterminable area itself and the region, edge noise position can determine, and will
Edge noise is changed to background pixel, thereby eliminates the influence that edge noise is divided crown word number and identified.Meanwhile the program
Handled for predeterminable area, and it is indirect whole image is handled, time complexity is low, and treatment effeciency is high.
Detailed description of the invention
Fig. 1 is the flow chart for the crown word number image processing method that the embodiment of the present invention one provides;
Fig. 2 is the structural block diagram for the crown word number image processing apparatus that the embodiment of the present invention three provides;
Fig. 3 is the structural schematic diagram for the equipment that the embodiment of the present invention four provides;
Fig. 4 is the part grayscale image schematic diagram for the bank note that the embodiment of the present invention six provides;
Fig. 5 is the grayscale image schematic diagram in the crown word number region that the embodiment of the present invention six provides;
Fig. 6 is the bianry image schematic diagrames in the crown word number region that the embodiment of the present invention six provides;
Fig. 7 is the predeterminable area and connected domain schematic diagram that the embodiment of the present invention six provides;
Fig. 8 is the updated bianry image schematic diagrames that the embodiment of the present invention six provides.
Specific embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention rather than limiting the invention.It also should be noted that in order to just
Only the parts related to the present invention are shown in description, attached drawing rather than entire infrastructure.
Embodiment one
Fig. 1 is the flow chart for the crown word number image processing method that the embodiment of the present invention one provides, and the present embodiment is applicable to
The case where there are marginal foreground points (also referred to as edge noise) in crown word number region, marginal foreground point refers to be removed in crown word number region
The foreground point in image border except crown word number, marginal foreground point are not connect with prefix sign character generally, marginal foreground
Point may will affect crown word number segmentation and identification, for example, its due to being truncated to when bank note feature itself intercepts crown word number region
His pattern, the foreground point in the upper left corner is marginal foreground point in image as shown in Figure 5.This method can be by crown word number image procossing
Device executes, and can be with calculating and the device of processing function, such as computer etc..As shown in Figure 1, this method is specifically wrapped
Include following steps:
Step 110, the bianry image in crown word number region is obtained.
In this step, the grayscale image in crown word number region can be intercepted from bank note grayscale image, to the gray scale in crown word number region
Figure carries out binaryzation, obtains the bianry image in crown word number region.Specifically, being known that crown word number according to the characteristics of bank note itself
Position Approximate on bank note grayscale image, it is possible thereby to intercept the grayscale image in crown word number region from bank note grayscale image.It can adopt
Binaryzation is carried out with existing Binarization methods, for example, Two-peak method, P parametric method, Otsu method, maximum entropy threshold method, iterative method
(best threshold method) etc., the embodiment of the present invention is to binarization without being described in detail.
Step 120, connected domain analysis is carried out to foreground pixel point each in the predeterminable area of bianry image, obtains predeterminable area
Connected domain.
In this step, predeterminable area refers to may be comprising the part region of edge noise, predeterminable area in bianry image
Position and size can be configured according to bank note actual conditions (such as edge noise position), for example, for certain currency type and/
Or the bank note of face amount, the position that edge noise occurs is usually fixed and does not connect with prefix sign character, then can
Subsequent processing is carried out so that the region comprising the position and partial character is arranged as predeterminable area.The embodiment of the present invention is to default
The shape in region is not construed as limiting, and may generally be rectangle, square etc..Connected domain, which refers to, has same pixel value and position in image
The image-region of adjacent foreground pixel point composition.Connected domain analysis refers to by row or by all prospect pictures of column access images
Vegetarian refreshments is found out all connected domains and is marked.Specifically, can be using eight connected region or four connected region to bianry image
Each foreground pixel point carries out connected domain analysis in predeterminable area.Connected domain analysis process and identical in the prior art, present invention reality
Example is applied to this without being described in detail.
Step 130, according to the number of connected domain and the feature of image of predeterminable area, the noise connection in connected domain is determined
Domain.
In this step, feature of image may include: edge noise be typically in image edge location, each word in crown word number
Symbol can regard that a connected domain, edge noise and character do not connect, predeterminable area includes at least partial character as.In conjunction with
The feature of image and connected domain number of predeterminable area itself can determine whether the predeterminable area includes noise, if comprising
Noise can determine which connected domain is noise connected domain, that is to say, that the position of edge noise is determined in bianry image.
Step 140, the pixel in noise connected domain is changed to background pixel point, obtains updated image.
In this step, connected domain is to change the pixel in noise connected domain for the foreground pixel in bianry image
For background pixel point, the value of the foreground pixel in noise connected domain is exactly changed to the corresponding value of background pixel, for example, two-value
Background pixel value is 0 in image, and foreground pixel value is 1, and the value of the pixel in noise connected domain is changed to 0, is realized side
Edge noise is set as background, does not include edge noise in thus obtained image.
The crown word number image processing method of the present embodiment, using connected domain analysis, the image in conjunction with predeterminable area itself is special
The number of connected domain, can determine edge noise position, and edge noise is changed to background pixel in point and the region, by
This eliminates the influence that edge noise is divided crown word number and identified.Meanwhile this method is handled for predeterminable area, rather than
Directly whole image is handled, time complexity is low, and treatment effeciency is high.
Preferably, prior to step 110, predeterminable area can be installed according to noise bits, wherein predeterminable area at least wraps
Include partial character.Crown word number region when for interception image not comprising edge noise, predeterminable area only include partial character;It is right
Crown word number region when interception image comprising edge noise, predeterminable area include edge noise and partial character.
For the bank note of same currency type and/or face amount, generally in interception image, the edge noise position basic one of interception
It causes.Specifically, the edge noise position being truncated to can be easy by test statistics, then set in advance according to edge noise position
Set position, the size and shape of predeterminable area.Corresponding edge noise position can also be determined according to the currency type and/or face amount of bank note
It sets, in practical applications, for the bank note of different currency type and/or face amount, correspondence can be pre-set by input unit
Predeterminable area, store its corresponding relationship, bank note examine when, according to currency type and/or value of money, read corresponding predeterminable area,
The determination and elimination of edge noise are carried out, the segmentation and identification of crown word number are then carried out.Alternatively, in use, for working as
Before the currency type and/or face amount to be examined, corresponding predeterminable area once is reset by input unit, executes acquisition again later
The step of image, image procossing and crown word number segmentation identification.
In above-mentioned preferred embodiment, the flexibility of method can be increased with self-setting predeterminable area, improve and eliminate side
The accuracy of edge noise.
By taking cash inspecting machine as an example, cash inspecting machine can receive many bank note, and presumable in received bank note there are edge noises, have
Received all bank note are executed described in the present embodiment according to predeterminable area during processing there is no edge noise
Method the paper money can determine according to the number of connected domain in the feature of image of predeterminable area itself and the region
Number region whether there is edge noise, if there is edge noise, it is determined that the position of edge noise, and edge noise is changed
For background pixel, the influence that edge noise is divided crown word number and identified is eliminated.
Embodiment two
The present embodiment is on the basis of the above embodiment 1, it is preferred that step 130 may comprise steps of: if even
The number in logical domain is one, it is determined that noise connected domain is not present;If the number of connected domain is two and more than two, right
Predeterminable area extends the pixel of predetermined number, connected domain analysis is carried out to the region after extension, according to new connected domain analysis knot
Fruit determines noise connected domain.
Since predeterminable area includes at least partial character, for the crown word number region of not edge noise, obtained connection
Domain should be the corresponding connected domain of partial character, therefore, if an only connected domain, indicates that the predeterminable area does not have edge to make an uproar
Sound.If the number of connected domain is two and two or more, predeterminable area is extended, connected domain analysis is carried out again, due to single
Character is connection, if new connected domain analysis result indicates former connected domain, there are intersections, and there are these of intersection
Connected domain may be character, rather than edge noise, it is possible thereby to determine that the former connected domain of no any intersection corresponds to side
The connected domain of edge noise.Wherein, the pixel of predetermined number, which can according to need, is configured.The present invention does not make the direction of extension
It limits, by taking Fig. 5 as an example, if edge noise is in the position between character 3 and 4 close to image border, predeterminable area is carried out
When extension, it can be extended along left side, right side and lower section.
Preferably, determine that noise connected domain may include following situations according to new connected domain analysis result:
(1) in the case where the number of former connected domain is two, if the number of new connected domain is one, it is determined that do not deposit
In noise connected domain;If the number of new connected domain is two, noise connected domain is determined according to the feature of image of predeterminable area.
For example, edge noise as shown in Figure 5, the position of predeterminable area is the upper left corner, to the predeterminable area after extension into
Row connected domain analysis obtains two new connected domains, is in image border according to edge noise and does not connect with character, single word
The characteristics of symbol is a connected domain, the connected domain marked as 1 can be determined as noise connected domain;For another example, edge noise is located at hat
The lower right corner in font size region, predeterminable area are the regions that lower right corner size is 20 × 20, if the number of new connected domain is two,
Then according to feature of image, the connected domain marked as 2 can be determined as noise connected domain.
(2) in the case where the number of former connected domain is three and three or more, former connected domain is judged according to new connected domain
Intersection situation.If there is intersection, it is determined that the connected domain for not having intersection in former connected domain is noise connected domain.If do not deposited
In intersection, then noise connected domain is determined according to the feature of image of predeterminable area, for example, carrying out connected domain point after continuing extended area
Analysis judges intersection, alternatively, determining that the connected domain close to predeterminable area edge is that noise connected domain determines label 1 by taking Fig. 7 as an example
Connected domain with 3 is noise connected domain.
The method that the above-mentioned feature of image according to connected domain number and predeterminable area determines edge noise is realized simple.
Embodiment three
Fig. 2 is the structural block diagram for the crown word number image processing apparatus that the embodiment of the present invention three provides, and the present embodiment is applicable
In there are edge noises in crown word number region the case where, as shown in Fig. 2, the device includes: image collection module 210, connected domain point
Analyse module 220, noise connected domain determining module 230 and image update module 240.
Image collection module 210, for obtaining the bianry image in crown word number region;
Connected domain analysis module 220 carries out connected domain minute for foreground pixel point each in the predeterminable area to bianry image
Analysis, obtains the connected domain of predeterminable area;
Noise connected domain determining module 230, for according to the number of connected domain, the feature of image of predeterminable area and default rule
Then, the noise connected domain in connected domain is determined;
Image update module 240, for the pixel in noise connected domain to be changed to background pixel point, after obtaining update
Image.
The crown word number image processing apparatus of the present embodiment, using connected domain analysis, the image in conjunction with predeterminable area itself is special
The number of connected domain, can determine edge noise position, and edge noise is changed to background pixel in point and the region, by
This eliminates the influence that edge noise is divided crown word number and identified.Meanwhile the device is handled for predeterminable area, rather than
Directly whole image is handled, time complexity is low, and treatment effeciency is high.
Preferably, above-mentioned apparatus can also include: predeterminable area setup module, for installing preset areas according to noise bits
Domain, wherein predeterminable area includes at least partial character.
Noise connected domain determining module 230 may include: the first connected domain determination unit and the second connected domain determination unit,
Wherein, in the case that the first connected domain determination unit for the number in connected domain is one, determine that there is no noises to be connected to
Domain;In the case that second connected domain determination unit for the number in connected domain is two and two or more, predeterminable area is expanded
The pixel for opening up predetermined number carries out connected domain analysis to the region after extension, determines noise according to new connected domain analysis result
Connected domain.
Second connected domain determination unit is specifically used for: in the case where the number of former connected domain is two, if new connection
The number in domain is two, then determines noise connected domain according to the feature of image of predeterminable area;If the number of new connected domain is one
It is a, it is determined that noise connected domain is not present;In the case where the number of former connected domain is three and three or more, according to new connection
Domain judges the intersection situation of former connected domain;If there is intersection, it is determined that the connected domain for not having intersection in former connected domain is noise
Connected domain;If there is no intersection, then noise connected domain is determined according to the feature of image of predeterminable area.
Preferably, connected domain analysis module 220 specifically can be using eight connected region or four connected region to bianry image
Each foreground pixel point carries out connected domain analysis in predeterminable area.
Image collection module 210 is specifically used for: the grayscale image in crown word number region is intercepted from bank note grayscale image;To crown word number
The grayscale image in region carries out binaryzation, obtains the bianry image in crown word number region.
Crown word number image processing method provided by any embodiment of the invention can be performed in above-mentioned crown word number image processing apparatus
Method has and executes the corresponding functional module of this method and beneficial effect.
Example IV
Fig. 3 is the structural schematic diagram for the equipment that the embodiment of the present invention four provides, as shown in figure 3, the equipment includes: processor
310, memory 320, input unit 330 and output device 340.The quantity of processor 310 can be one or more in the equipment
It is a, in Fig. 3 by taking a processor 310 as an example;Processor 310, memory 320, input unit 330 and output device in the equipment
340 can be connected by bus or other modes, in Fig. 3 for being connected by bus.
Memory 320 is used as a kind of computer readable storage medium, can be used for storing software program, journey can be performed in computer
Sequence and module, if the corresponding program instruction/module of crown word number image processing method in the embodiment of the present invention is (for example, prefix
Image collection module 210, connected domain analysis module 220, noise connected domain determining module 230 and figure in number image processing apparatus
As update module 240).Software program, instruction and the module that processor 310 is stored in memory 320 by operation, thus
The various function application and data processing of equipment are executed, that is, realizes the crown word number image processing method of above-described embodiment one and two
Method.
Memory 320 can mainly include storing program area and storage data area, wherein storing program area can store operation system
Application program needed for system, at least one function;Storage data area, which can be stored, uses created data etc. according to terminal.This
Outside, memory 320 may include high-speed random access memory, can also include nonvolatile memory, for example, at least one
Disk memory, flush memory device or other non-volatile solid state memory parts.In some instances, memory 320 can be into one
Step includes the memory remotely located relative to processor 310, these remote memories can pass through network connection to equipment.On
The example for stating network includes but is not limited to internet, intranet, local area network, mobile radio communication and combinations thereof.
Input unit 330 can be used for receiving the information of input, and generate the user setting and function control with equipment
Related key signals input, for example, receiving the information that user is arranged about predeterminable area.Output device 340 may include display screen
Deng display equipment, for example, display grayscale image, bianry image, crown word number recognition result etc..Above equipment can also be adopted including image
Acquisition means, for acquiring banknote image.
Embodiment five
The embodiment of the present invention five provides a kind of computer readable storage medium, is stored thereon with computer program, the journey
The crown word number image processing method as described in embodiment one and two is realized when sequence is executed by processor.
By the description above with respect to embodiment, it is apparent to those skilled in the art that, the present invention
It can be realized by software and required common hardware, naturally it is also possible to which by hardware realization, but in many cases, the former is more
Good embodiment.Based on this understanding, technical solution of the present invention substantially in other words contributes to the prior art
Part can be embodied in the form of software products, which can store in computer readable storage medium
In, floppy disk, read-only memory (Read-Only Memory, ROM), random access memory (Random such as computer
Access Memory, RAM), flash memory (FLASH), hard disk or CD etc., including some instructions are with so that a computer is set
Standby (can be personal computer, server or the network equipment etc.) executes method described in each embodiment of the present invention.
It is worth noting that, in the embodiment of above-mentioned crown word number image processing apparatus, included each unit and module
It is only divided according to the functional logic, but is not limited to the above division, as long as corresponding functions can be realized;
In addition, the specific name of each functional unit is also only for convenience of distinguishing each other, the protection scope being not intended to restrict the invention.
Embodiment six
The present embodiment provides a kind of specific example, with reference to the accompanying drawing to hat based on above-described embodiment one to five
Font size image processing process is illustrated.
As shown in figure 4, nearby there are other patterns for crown word number, it can according to the position of banknote image and crown word number on bank note
To intercept the grayscale image in crown word number region from Fig. 4, as shown in Figure 5.Two are carried out to the grayscale image in crown word number region shown in fig. 5
Value obtains bianry image shown in fig. 6, it is clear that there are edge noises in the image upper left corner, may will affect crown word number segmentation.
The predeterminable area being arranged in advance is the square area that the size since the bianry image upper left corner is 12 × 12, such as
Shown in Fig. 7, connected domain analysis is carried out to predeterminable area using eight connected region, obtains three connected domains, label is respectively 1,2,3.
Predeterminable area is extended into N number of pixel, for example, 5 pixels of extension obtain 17 × 17 region, 17 × 17 region is carried out
Connected domain analysis, obtains two new connected domains, indicates that pixel and the pixel marked as 3 marked as 2 have an intersection, i.e., this
Two connected domains are two parts in character " 3 ", it is possible thereby to determine that the pixel marked as 1 is edge noise, by label
Background dot is labeled as 1 pixel.The image re-flagged is as shown in figure 8, eliminate marginal foreground point.If new company
Logical domain analysis result is still without intersection, according to the characteristic in crown word number region itself, can will be close to image border marked as 1
Pixel with 3 is denoted as background dot.
Note that the above is only a better embodiment of the present invention and the applied technical principle.It will be appreciated by those skilled in the art that
The invention is not limited to the specific embodiments described herein, be able to carry out for a person skilled in the art it is various it is apparent variation,
It readjusts and substitutes without departing from protection scope of the present invention.Therefore, although being carried out by above embodiments to the present invention
It is described in further detail, but the present invention is not limited to the above embodiments only, without departing from the inventive concept, also
It may include more other equivalent embodiments, and the scope of the invention is determined by the scope of the appended claims.
Claims (10)
1. a kind of crown word number image processing method characterized by comprising
Obtain the bianry image in crown word number region;
Connected domain analysis is carried out to foreground pixel point each in the predeterminable area of the bianry image, obtains the company of the predeterminable area
Logical domain;
According to the number of connected domain and the feature of image of the predeterminable area, the noise connected domain in the connected domain is determined;
Pixel in the noise connected domain is changed to background pixel point, obtains updated image.
2. the method according to claim 1, wherein obtain crown word number region bianry image before, it is described
Method further include:
The predeterminable area is installed according to noise bits, wherein the predeterminable area includes at least partial character.
3. the method according to claim 1, wherein according to the number of connected domain and the image of the predeterminable area
Feature determines the noise connected domain in the connected domain, comprising:
If the number of connected domain is one, it is determined that noise connected domain is not present;
If the number of connected domain is two and two or more, to the pixel of predeterminable area extension predetermined number, to expansion
Region after exhibition carries out connected domain analysis, determines noise connected domain according to new connected domain analysis result.
4. according to the method described in claim 3, it is characterized in that, determining that noise is connected to according to new connected domain analysis result
Domain, comprising:
In the case where the number of former connected domain is two, if the number of new connected domain is one, it is determined that noise is not present
Connected domain;If the number of new connected domain is two, noise connected domain is determined according to the feature of image of predeterminable area;
In the case where the number of former connected domain is three and three or more, the friendship of the former connected domain is judged according to new connected domain
Collect situation;If there is intersection, it is determined that the connected domain for not having intersection in former connected domain is noise connected domain;If there is no friendship
Collection, then determine noise connected domain according to the feature of image of the predeterminable area.
5. the method according to claim 1, wherein to each foreground pixel in the predeterminable area of the bianry image
Point carries out connected domain analysis, comprising:
Connected domain is carried out to foreground pixel point each in the predeterminable area of the bianry image using eight connected region or four connected region
Analysis.
6. the method according to claim 1, wherein obtaining the bianry image in crown word number region, comprising:
The grayscale image in crown word number region is intercepted from bank note grayscale image;
Binaryzation is carried out to the grayscale image in the crown word number region, obtains the bianry image in the crown word number region.
7. a kind of crown word number image processing apparatus characterized by comprising
Image collection module, for obtaining the bianry image in crown word number region;
Connected domain analysis module carries out connected domain analysis for foreground pixel point each in the predeterminable area to the bianry image,
Obtain the connected domain of the predeterminable area;
Noise connected domain determining module, for number, the feature of image and preset rules of the predeterminable area according to connected domain,
Determine the noise connected domain in the connected domain;
Image update module obtains updated for the pixel in the noise connected domain to be changed to background pixel point
Image.
8. device according to claim 7, which is characterized in that the noise connected domain determining module includes:
In the case that first connected domain determination unit for the number in connected domain is one, determines and noise connected domain is not present;
In the case that second connected domain determination unit for the number in connected domain is two and two or more, to described default
Region extends the pixel of predetermined number, carries out connected domain analysis to the region after extension, true according to new connected domain analysis result
Determine noise connected domain.
9. a kind of equipment, which is characterized in that the equipment includes:
One or more processors;
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as crown word number image processing method described in any one of claims 1 to 6.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Such as crown word number image processing method described in any one of claims 1 to 6 is realized when execution.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710384027.9A CN108961532B (en) | 2017-05-26 | 2017-05-26 | Method, device and equipment for processing crown word number image and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710384027.9A CN108961532B (en) | 2017-05-26 | 2017-05-26 | Method, device and equipment for processing crown word number image and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108961532A true CN108961532A (en) | 2018-12-07 |
CN108961532B CN108961532B (en) | 2020-11-17 |
Family
ID=64494309
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710384027.9A Active CN108961532B (en) | 2017-05-26 | 2017-05-26 | Method, device and equipment for processing crown word number image and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108961532B (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102194118A (en) * | 2010-03-02 | 2011-09-21 | 方正国际软件(北京)有限公司 | Method and device for extracting information from image |
CN104268978A (en) * | 2014-09-24 | 2015-01-07 | 苏州佳世达光电有限公司 | Method for identifying serial number on bill |
US20150109290A1 (en) * | 2013-10-22 | 2015-04-23 | Fu Tai Hua Industry (Shenzhen) Co., Ltd. | Device and method for removing noise points in point clouds |
US20150371399A1 (en) * | 2014-06-19 | 2015-12-24 | Kabushiki Kaisha Toshiba | Character Detection Apparatus and Method |
US20150379338A1 (en) * | 2014-06-30 | 2015-12-31 | Lg Cns Co., Ltd. | Apparatus and Method for Recognizing Media and Financial Device |
CN105303533A (en) * | 2015-11-03 | 2016-02-03 | 华中科技大学 | Ultrasonic image filtering method |
CN105913093A (en) * | 2016-05-03 | 2016-08-31 | 电子科技大学 | Template matching method for character recognizing and processing |
CN106023204A (en) * | 2016-05-20 | 2016-10-12 | 陕西师范大学 | Method and system for removing mosquito noise based on edge detection algorithm |
-
2017
- 2017-05-26 CN CN201710384027.9A patent/CN108961532B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102194118A (en) * | 2010-03-02 | 2011-09-21 | 方正国际软件(北京)有限公司 | Method and device for extracting information from image |
US20150109290A1 (en) * | 2013-10-22 | 2015-04-23 | Fu Tai Hua Industry (Shenzhen) Co., Ltd. | Device and method for removing noise points in point clouds |
US20150371399A1 (en) * | 2014-06-19 | 2015-12-24 | Kabushiki Kaisha Toshiba | Character Detection Apparatus and Method |
US20150379338A1 (en) * | 2014-06-30 | 2015-12-31 | Lg Cns Co., Ltd. | Apparatus and Method for Recognizing Media and Financial Device |
CN104268978A (en) * | 2014-09-24 | 2015-01-07 | 苏州佳世达光电有限公司 | Method for identifying serial number on bill |
CN105303533A (en) * | 2015-11-03 | 2016-02-03 | 华中科技大学 | Ultrasonic image filtering method |
CN105913093A (en) * | 2016-05-03 | 2016-08-31 | 电子科技大学 | Template matching method for character recognizing and processing |
CN106023204A (en) * | 2016-05-20 | 2016-10-12 | 陕西师范大学 | Method and system for removing mosquito noise based on edge detection algorithm |
Non-Patent Citations (2)
Title |
---|
孟庆远: "基于字符连通域特征的车牌", 《测试技术学报》 * |
郭俊芳: "纸币面额与序列号识别算法的设计与实现", 《北京邮电大学》 * |
Also Published As
Publication number | Publication date |
---|---|
CN108961532B (en) | 2020-11-17 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10885644B2 (en) | Detecting specified image identifiers on objects | |
CN104751187B (en) | Meter reading automatic distinguishing method for image | |
CN107437294B (en) | Character segmentation method, device, equipment and storage medium | |
CN109685055A (en) | Text filed detection method and device in a kind of image | |
CN104867225A (en) | Banknote face orientation identification method and apparatus | |
US10747993B2 (en) | Detecting a fragmented object in an image | |
CN106952393B (en) | Paper money identification method and device, electronic equipment and storage medium | |
CN110443159A (en) | Digit recognition method, device, electronic equipment and storage medium | |
CN104751093A (en) | Method and device for acquiring image identification code displayed by host equipment | |
CN112598627A (en) | Method, system, electronic device and medium for detecting image defects | |
CN109035256A (en) | User interface image cutting method, device, server and storage medium | |
CN106898078B (en) | Port currency version identification method and device | |
CN109087439B (en) | Bill checking method, terminal device, storage medium and electronic device | |
CN110909816B (en) | Picture identification method and device | |
CN106683257A (en) | Serial number location method and device | |
CN108961532A (en) | Crown word number image processing method, device, equipment and storage medium | |
CN111008635A (en) | OCR-based multi-bill automatic identification method and system | |
CN112348112B (en) | Training method and training device for image recognition model and terminal equipment | |
CN113449686A (en) | Identification method, device, equipment and medium for identity card counterfeiting | |
CN109582408B (en) | Data display method, device and equipment | |
CN108961531B (en) | Method, device and equipment for identifying serial number of paper currency and storage medium | |
CN111626244A (en) | Image recognition method, image recognition device, electronic equipment and medium | |
US10223618B1 (en) | Method and apparatus for transformation of dot text in an image into stroked characters based on dot pitches | |
US10192132B1 (en) | Method and apparatus for detection of dots in an image | |
CN111599080A (en) | Spliced paper money detection method and device, financial machine tool equipment and storage medium |
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 |