CN110135218A - The method, apparatus, equipment and computer storage medium of image for identification - Google Patents
The method, apparatus, equipment and computer storage medium of image for identification Download PDFInfo
- Publication number
- CN110135218A CN110135218A CN201810105171.9A CN201810105171A CN110135218A CN 110135218 A CN110135218 A CN 110135218A CN 201810105171 A CN201810105171 A CN 201810105171A CN 110135218 A CN110135218 A CN 110135218A
- Authority
- CN
- China
- Prior art keywords
- module
- form image
- image
- region
- extreme point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 230000009467 reduction Effects 0.000 claims description 17
- 238000003708 edge detection Methods 0.000 claims description 16
- 238000012545 processing Methods 0.000 claims description 12
- 230000009466 transformation Effects 0.000 claims description 8
- 238000004590 computer program Methods 0.000 claims description 5
- 238000010586 diagram Methods 0.000 description 15
- 238000004891 communication Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012015 optical character recognition Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/48—Extraction of image or video features by mapping characteristic values of the pattern into a parameter space, e.g. Hough transformation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/40—Document-oriented image-based pattern recognition
- G06V30/41—Analysis of document content
- G06V30/412—Layout analysis of documents structured with printed lines or input boxes, e.g. business forms or tables
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
- G06T2207/10008—Still image; Photographic image from scanner, fax or copier
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20182—Noise reduction or smoothing in the temporal domain; Spatio-temporal filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30176—Document
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Abstract
According to the example implementations of present disclosure, the method, apparatus, equipment and computer storage medium of a kind of image for identification are provided.Specifically, a kind of method of image for identification is provided, comprising: obtain form image, which includes table line and table content, and the table line is associated with multiple fields in the table content;Based on the form image, the table line is identified;Based on the table line identified, which is divided into multiple regions;And based on the multiple region divided, identify the content of multiple field.According to the example implementations of present disclosure, the corresponding device of image for identification, equipment and computer storage medium are additionally provided.
Description
Summary of the invention
Embodiment of the disclosure relates in general to image recognition, and in particular to the method for the table line in identification form image,
Device, equipment and computer storage medium.
Summary of the invention
Table is the visual pattern of group organization data.For a long time, be widely used the table of various paper-based forms with
Support the activities such as communication exchanges, scientific research and data analysis.For example, such as balance sheet, cash flow statement, profit flow table
Deng enterprise financial report can be used to indicate that enterprise in the management state of specific period, and can be used for mentioning to financial institution
For the decision-making foundation of credit examination & approval.With the development of office digitizing technique, need existing paper list being converted to electronics
The form of table.Therefore, it in order to more accurately convey information included in table, needs to be modified to identify form image
Technology.
Summary of the invention
According to the example embodiment of present disclosure, a kind of scheme of image for identification is provided.
In the first aspect of present disclosure, a kind of method of image for identification is provided.Specifically, this method packet
It includes: obtaining form image, which includes table line and table content, multiple words in the table line and the table content
Section is associated;Based on the form image, the table line is identified;Based on the table line identified, which is divided into more
A region;And based on the multiple region divided, identify the content of multiple field.
In in the second aspect of the present disclosure, a kind of device of image for identification is provided.Specifically, the device packet
It includes: obtaining module, which is configured as obtaining form image, which includes table line and table content, is somebody's turn to do
Table line is associated with multiple fields in the table content;Table line identification module, the table line identification module are configured as
Based on the form image, the table line is identified;Division module, the division module are configured as based on the table line identified, will
The form image is divided into multiple regions;And field identification module, the field identification module are configured as based on being divided
Multiple region identifies the content of multiple field.
In the third aspect of present disclosure, a kind of equipment, including one or more processors are provided;And storage
Device, for storing one or more programs, when one or more programs are executed by one or more processors so that one or
The method that multiple processors realize the first aspect according to present disclosure.
In the fourth aspect of present disclosure, a kind of computer-readable medium is provided, is stored thereon with computer journey
Sequence realizes the method for the first aspect according to present disclosure when the program is executed by processor.
It should be appreciated that content described in Summary is not intended to limit the pass of the embodiment of present disclosure
Key or important feature, it is also non-for limiting the scope of the disclosure.The other feature of present disclosure will be retouched by below
It states and is easy to understand.
Detailed description of the invention
It refers to the following detailed description in conjunction with the accompanying drawings, the above and other feature, advantage of each embodiment of present disclosure
And aspect will be apparent.In the accompanying drawings, the same or similar appended drawing reference indicates the same or similar element, in which:
Fig. 1 is shown the schematic diagram of processed form image;
Fig. 2 shows the flow charts according to the method for the image for identification of the exemplary embodiment of present disclosure;
Fig. 3 shows the signal of the hough space for form image of the exemplary embodiment according to present disclosure
Figure;
Fig. 4 shows the schematic diagram of the extreme point in the hough space according to the exemplary embodiment of present disclosure;
Fig. 5 shows the schematic diagram for the table line of the exemplary embodiment according to present disclosure identified;
Fig. 6 shows the form image with identified table line of the exemplary embodiment according to present disclosure
Schematic diagram;
Fig. 7 diagrammatically illustrates the device of the form image for identification of the exemplary embodiment according to present disclosure
Block diagram;And
Fig. 8 shows the block diagram that can implement the calculating equipment of multiple embodiments of present disclosure.
Specific embodiment
The embodiment of present disclosure is more fully described below with reference to accompanying drawings.Although being shown in the disclosure in attached drawing
The some embodiments of appearance, it should be understood that, present disclosure can be realized by various forms, and should not be by
It is interpreted as being limited to embodiments set forth here, providing these embodiments on the contrary is in order to more thorough and be fully understood by the disclosure
Content.It should be understood that the being given for example only property of accompanying drawings and embodiments of present disclosure acts on, it is not intended to limit the disclosure
The protection scope of content.
In the description of the embodiment of present disclosure, term " includes " and its similar term should be understood as open packet
Contain, i.e., " including but not limited to ".Term "based" should be understood as " being based at least partially on ".Term " one embodiment " or
" embodiment " should be understood as " at least one embodiment ".Term " first ", " second " etc. may refer to different or phase
Same object.Hereafter it is also possible that other specific and implicit definition.
Traditionally, form image can pass through such as optical character identification (Optical Character
Recognition, OCR) and it is converted into the table of computer processable form, for such as word processor, electronic watch
The software of lattice software etc. is further processed.However, the table line in form image is easy so that table content is unclear, thus difficult
To accurately identify form image.
For example, Fig. 1 is shown the schematic diagram of processed form image 100.Form image 100 includes multiple field groups
1101-110N, wherein N is the natural number greater than 1.Field groups 1101-110NRespectively include multiple fields, and each field packet
Include corresponding field contents.For example, field groups 1101Including field 11011-1101M, and field groups 110NIncluding field
110N1-110NK, wherein M and K is the natural number greater than 1.Although by field 1101 in Fig. 1MAnd 110NKIt is shown as alignment (i.e.
M=K), however, it is understood that in other embodiments, field 1101MIt can be with field 110NKIt is misaligned (i.e. M ≠ K).In addition, also
It should be understood that field contents can be sky.
As shown in Figure 1, in form image to be processed may there are following situations, the table line in form image 100
With field 11011With 11012In field contents overlapping so that field 11011With 11012It is unclear.At this point, table turns
System is changed when form image 100 to be converted to the table of computer processable form, it may be by field 11011With 11012Word
Section content is all identified as field 11012Field contents, or there are other mistakes.Further, table line is also easy to make
Identified field contents inaccuracy is obtained, to reduce the accuracy of identification form image.
For one or more problem at least being partially solved in the above problem and other potential problems, the disclosure
Example embodiment propose a kind of scheme of image for identification.Table content of the program in identification form image 100
The table line in form image 100 is identified before, is divided into form image 100 and table content based on the table line identified
In the corresponding multiple regions of multiple fields, the content of field is then identified by the content in identification region.With this side
Formula, though in the case where table line does not know table content, can also it is more quick, accurately and effectively identify table
Table content in image 100, to improve the accuracy of table converting system.
Fig. 2 shows the flow charts according to the method 200 of the image for identification of the exemplary embodiment of present disclosure.
For example, this method 200 can be executed by table converting system.It should be understood that this method 200 can also include being not shown
Additional step and/or can be omitted shown step, the scope of the present disclosure is not limited in this respect.
210, table converting system obtains form image 100.In certain embodiments, table converting system can be from all
The various equipment that such as camera, scanner, smart phone can be used for obtaining the image of table obtain form image 100.For example,
Table converting system can obtain the form image for being directed to enterprise financial report from scanner.Form image 100 may include table
Ruling and table content.Table line is associated with multiple fields in table content.For example, multiple fields in table content can
To be distinguished by table line.
220, table converting system can identify the table in form image 100 based on acquired form image 100
Line.In certain embodiments, table converting system can carry out binaryzation to form image 100 to generate bianry image.
In certain embodiments, table converting system can carry out gray processing to form image 100 to generate through gray processing
Form image.The color of form image can be removed by carrying out gray processing to form image 100, and only retain the bright of form image
Degree.Then, table converting system can carry out binaryzation to the form image through gray processing to generate bianry image.For example, table
Pixel in form image through gray processing can be converted to gray value based on predetermined gray threshold and only be set by lattice converting system
It is set to such as 0 or 255 pixel, to generate bianry image.
In certain embodiments, table converting system can carry out noise reduction to form image 100 to generate the table through noise reduction
Then table images carry out binaryzation to the form image through noise reduction to generate bianry image.Carrying out noise reduction to form image 100 can
To remove the noise in form image 100, so that form image 100 is more smooth.For example, table converting system can pass through height
This filtering, mean filter, deep learning scheduling algorithm realize noise reduction.
Alternatively, table converting system can be to the application edge detection of form image 100 to generate the table through edge detection
Table images, and binaryzation then is carried out to generate bianry image to the form image through edge detection.To form image 100 into
Row edge detection can retain the important feature attribute of the table line in form image 100, remove inessential or incoherent letter
Breath, so that form image 100 is more suitable for detecting table line.
Table converting system can carry out Hough transformation to bianry image generated and be directed to bianry image suddenly to generate
Husband space.Hough space is the space opposite with the real space of bianry image.
In certain embodiments, real space can be indicated with cartesian coordinate system.In cartesian coordinate system, binary map
Straight line as in can be determined by two points in cartesian coordinate system, and the straight line can have slope and intercept.
It, can be using the slope of straight line and intercept as the reference axis of hough space when generating the hough space for being directed to bianry image.With
This mode, the straight line in cartesian coordinate system can be represented as a point in hough space, and cartesian coordinate
A point in system can correspondingly be represented as the straight line in hough space.
Alternatively, in certain embodiments, real space can also be indicated with polar coordinate system.In such a case, it is possible to
Using polar diameter ρ and polar angle θ as the reference axis of hough space.By this method, the straight line in polar coordinate system can be represented as
A point in hough space, and a point in polar coordinate system can be represented as a curve in hough space.Fig. 3
Show the schematic diagram of the hough space 300 for form image 100 of the exemplary embodiment according to present disclosure.Such as figure
Shown, the vertical axis of hough space 300 is polar diameter ρ, and trunnion axis is polar angle θ.Furthermore, it is to be understood that a plurality of ash in Fig. 3
Color curve is expression of the pixel in hough space in bianry image.
Table converting system can determine the extreme point in the hough space for bianry image.In cartesian coordinate system
In, if multiple points (i.e. slope is identical with intercept) on same straight line, these points are corresponding a plurality of in hough space
Straight line intersection is in same point.Similarly, in polar coordinate system, if multiple points (i.e. polar diameter ρ and polar angle θ on same straight line
It is identical), then these points in hough space corresponding a plurality of curve intersection in same point.
Based on the above principles, since the table line in cartesian coordinate system/polar coordinate system in bianry image is multiple pictures
Vegetarian refreshments constitute straight line, therefore these pixels in hough space corresponding a plurality of straight line/curve intersection in same point.The friendship
Point indicates the table line.Further, since table line is usually longer straight line in form image 100, therefore table converting system
Table line can be determined by determining the extreme point in hough space in the intersection point of a plurality of straight line/curve.
In certain embodiments, the available predetermined threshold related with the feature of extreme point of table converting system, and
Extreme point is determined based on predetermined threshold.For example, this feature may include the number of extreme point.For example, being obtained in table converting system
In the case where the predetermined threshold for taking 6 extreme points, table converting system can be determined in hough space by the maximum number of straight line
6 intersection points of intersection are as extreme point.Alternatively, this feature may include the size of extreme point.For example, in table converting system
In the case where the predetermined threshold for obtaining the intersection point of 100 straight line intersections, table converting system can be determined in hough space at least
By the intersection point of 100 straight line intersections as extreme point.
Fig. 4 shows the schematic diagram of the extreme point in the hough space 300 according to the exemplary embodiment of present disclosure.
As shown, table converting system can determine multiple extreme points in hough space 300, such as extreme point 410.
Table converting system can determine table line based on identified extreme point.For example, table converting system can incite somebody to action
Extreme point in hough space 300 is converted to the straight line in real space, so that it is determined that the straight line is as table line.Fig. 5 is shown
According to the schematic diagram of the table line 500 that is identified of the exemplary embodiment of present disclosure.It is represented by dashed line and is known in figure
Other table line.As an example, the table line 510 identified is the straight line that the extreme point 410 in hough space 300 is converted.Class
As, based on other extreme points in hough space 300 as shown in Figure 4, it can also determine other table lines 500 in Fig. 5.
Form image 100 can be divided into multiple regions based on the table line identified by table converting system.Certain
In embodiment, table converting system can be based on coordinate of the table line identified in form image 100, by form image
100 are divided into multiple regions.What Fig. 6 was shown according to the exemplary embodiment of present disclosure with identified table line
The schematic diagram of form image 600.
In Fig. 6, actual table line is indicated by solid line, and the table line identified is represented by the dotted line.In order to become apparent from
Ground shows table line, is shown as separating with actual table line by the table line identified in Fig. 6, it should be understood that the table identified
Ruling can be with actual table line overlap.
As shown in fig. 6, the table line that form image 600 is identified is divided into multiple regions group 6101-610N, wherein N be
Natural number greater than 1.Region group 6101-610NRespectively include multiple regions.For example, region group 6101Including region 61011-
6101M, and region group 610NIncluding region 610N1-610NK.It is similar with field, although by region 6101 in Fig. 6MWith
610NKIt is shown as alignment (i.e. M=K), however, it is understood that in other embodiments, region 6101MIt can be with region 610NKIt is not right
(i.e. M ≠ K) together.
Each region in multiple regions corresponds to each field in multiple fields.For example, region 61011-6101MPoint
It Dui Yingyu not field 11011-1101M, and region 61011-6101KCorrespond respectively to field 11011-1101K。
Table converting system can identify the content of multiple fields based on the multiple regions divided.In some embodiments
In, for one of a plurality of areas region, table converting system can determine multiple words by identifying the content in the region
The content of the corresponding field in the region Duan Zhongyu.Table converting system can be handled form image 100 is converted to computer
After the table of form, it can store converted table and be provided for downloading.
By this method, form image 100 can be divided into area corresponding with field based on table line by table converting system
Domain, with the content based on region recognition corresponding field.In such a case, it is possible to form image on the basis of every field
100 are identified, and reduce interference of the table line to field contents, to improve the reject rate of table converting system, miss
Knowledge rate, recognition speed, stability, ease for use and feasibility etc..
Fig. 7 diagrammatically illustrates the dress of the form image for identification 100 of the exemplary embodiment according to present disclosure
Set 700 block diagram.Specifically, which includes: to obtain module 710, obtains module 710 and is configured as obtaining form image,
The form image includes table line and table content, and the table line is associated with multiple fields in the table content;Table line
Identification module 720, the table line identification module 720 are configured as identifying the table line based on the form image;Division module
730, which is configured as that the form image is divided into multiple regions based on the table line identified;And word
Section identification module 740, the field identification module 740 are configured as identifying multiple field based on the multiple region divided
Content.
In certain embodiments, table line identification module 720 includes: binarization block, which is configured as
Binaryzation is carried out to generate bianry image to the form image;Hough transformation module, the Hough transformation module are configured as to this
Bianry image carries out Hough transformation to generate the hough space for being directed to the bianry image;Extreme point determining module, the extreme point are true
Cover half block is configured to determine that the extreme point in the hough space;And table line determining module, the table line determining module quilt
It is configured to the extreme point, determines the table line.
In certain embodiments, binarization block includes: noise reduction module, which is configured as to the form image
Noise reduction is carried out to generate the form image through noise reduction;And the first bianry image generation module, first bianry image generate mould
Block is configured as the form image to this through noise reduction and carries out binaryzation to generate bianry image.Alternatively, binarization block includes:
Edge detection module, the edge detection module are configured as to the form image application edge detection to generate through edge detection
Form image;And the second bianry image generation module, the second bianry image generation module are configured as examining this through edge
The form image of survey carries out binaryzation to generate the bianry image.
In certain embodiments, extreme point determining module includes: that threshold value obtains module, which obtains module and be configured as
Predetermined threshold related with the feature of the extreme point is obtained, this feature includes in the number of the extreme point and the size of the extreme point
Any one;And threshold value extreme point determining module, the threshold value extreme point determining module are configured as true based on the predetermined threshold
The fixed extreme point.
In certain embodiments, division module 730 includes: region division module, which is configured as base
In the coordinate of the table line in the form image, which is divided into multiple region, it is every in multiple region
A region corresponds to each field in multiple field.
In certain embodiments, field identification module 740 includes: region identification module, which is configured
For for a region in multiple region, by identifying the content in the region, determine in multiple field with it is described
The content of the corresponding field in region.
According to the example implementations of present disclosure, a kind of equipment, including one or more processors are provided;With
And storage device, for storing one or more programs.When one or more programs are executed by one or more processors, make
One or more processors realization is obtained according to the method for present disclosure.
According to the example implementations of present disclosure, a kind of computer-readable medium is provided, is stored thereon with meter
Calculation machine program realizes the method according to present disclosure when the program is executed by processor.
Fig. 8 shows the block diagram that can implement the calculating equipment 800 of multiple embodiments of present disclosure.Equipment 800 can
For realizing table converting system.As shown, equipment 800 includes central processing unit (CPU) 801, it can be according to depositing
It stores up the computer program instructions in read-only memory (ROM) 802 or is loaded into random access storage device from storage unit 808
(RAM) computer program instructions in 803, to execute various movements appropriate and processing.In RAM 803, it can also store and set
Various programs and data needed for standby 800 operation.CPU 801, ROM 802 and RAM 803 are connected with each other by bus 804.
Input/output (I/O) interface 805 is also connected to bus 804.
Multiple components in equipment 800 are connected to I/O interface 805, comprising: input unit 806, such as keyboard, mouse etc.;
Output unit 807, such as various types of displays, loudspeaker etc.;Storage unit 808, such as disk, CD etc.;And it is logical
Believe unit 809, such as network interface card, modem, wireless communication transceiver etc..Communication unit 809 allows equipment 800 by such as
The computer network of internet and/or various telecommunication networks exchange information/data with other equipment.
Processing unit 801 executes each method as described above and processing, such as process 200.For example, in some implementations
In example, process 200 can be implemented as computer software programs, be tangibly embodied in machine readable media, such as storage list
Member 808.In some embodiments, some or all of of computer program can be via ROM 802 and/or communication unit 809
And it is loaded into and/or is installed in equipment 800.It, can be with when computer program loads to RAM 803 and when being executed by CPU 801
Execute the one or more steps of procedures described above 800 and/or process 900.Alternatively, in other embodiments, CPU
801 can be configured as implementation procedure 200 by other any modes (for example, by means of firmware) appropriate.
Function described herein can be executed at least partly by one or more hardware logic components.Example
Such as, without limitation, the hardware logic component for the exemplary type that can be used includes: field programmable gate array (FPGA), dedicated
Integrated circuit (ASIC), Application Specific Standard Product (ASSP), the system (SOC) of system on chip, load programmable logic device
(CPLD) etc..
Program code for implementing the method for present disclosure can be using any group of one or more programming languages
It closes to write.These program codes can be supplied to general purpose computer, special purpose computer or other programmable data processing units
Processor or controller so that program code when by processor or controller execution when make to be advised in flowchart and or block diagram
Fixed function/operation is carried out.Program code can be executed completely on machine, partly be executed on machine, as independence
Software package partly executes on machine and partly executes or hold on remote machine or server on the remote machine completely
Row.
In the context of present disclosure, machine readable media can be tangible medium, may include or stores
The program for using or being used in combination with instruction execution system, device or equipment for instruction execution system, device or equipment.Machine
Device readable medium can be machine-readable signal medium or machine-readable storage medium.Machine readable media may include but unlimited
In times of electronics, magnetic, optical, electromagnetism, infrared or semiconductor system, device or equipment or above content
What appropriate combination.The more specific example of machine readable storage medium will include the electrical connection of line based on one or more, portable
Formula computer disks, hard disk, random access memory (RAM), read-only memory (ROM), Erasable Programmable Read Only Memory EPROM
(EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage are set
Standby or above content any appropriate combination.
Although this should be understood as requiring operating in this way with shown in addition, depicting each operation using certain order
Certain order out executes in sequential order, or requires the operation of all diagrams that should be performed to obtain desired result.
Under certain environment, multitask and parallel processing be may be advantageous.Similarly, although containing several tools in being discussed above
Body realize details, but these be not construed as to scope of the present disclosure limitation.In the upper of individual embodiment
Certain features described in hereafter can also be realized in combination in single realize.On the contrary, in the context individually realized
The various features of description can also be realized individually or in any suitable subcombination in multiple realizations.
Although having used specific to this theme of the language description of structure feature and/or method logical action, answer
When understanding that theme defined in the appended claims is not necessarily limited to special characteristic described above or movement.On on the contrary,
Special characteristic described in face and movement are only to realize the exemplary forms of claims.
Claims (16)
1. a kind of method of image for identification, comprising:
Obtain form image, the form image includes table line and table content, in the table line and the table content
Multiple fields it is associated;
Based on the form image, the table line is identified;
Based on the table line identified, the form image is divided into multiple regions;And
Based on the multiple region divided, the content of the multiple field is identified.
2. according to the method described in claim 1, wherein identifying that the table line includes:
Binaryzation is carried out to generate bianry image to the form image;
Hough transformation is carried out to generate the hough space for being directed to the bianry image to the bianry image;
Determine the extreme point in the hough space;And
Based on the extreme point, the table line is determined.
3. according to the method described in claim 2, wherein generating the bianry image and including:
Noise reduction is carried out to generate the form image through noise reduction to the form image;And
Binaryzation is carried out to generate bianry image to the form image through noise reduction.
4. according to the method described in claim 2, wherein generating the bianry image and including:
To the form image application edge detection to generate the form image through edge detection;And
Binaryzation is carried out to generate the bianry image to the form image through edge detection.
5. according to the method described in claim 2, wherein determining that the extreme point includes:
Obtain related with the feature of extreme point predetermined threshold, the feature include the extreme point number and the pole
It is worth any one of the size of point;And
The extreme point is determined based on the predetermined threshold.
6. according to the method described in claim 1, the form image be wherein divided into multiple regions including:
The form image is divided into the multiple region, institute by the coordinate based on the table line in the form image
State each field that each region in multiple regions corresponds in the multiple field.
7. according to the method described in claim 6, wherein identifying that the content of the multiple field includes:
It is determined in the multiple field for a region in the multiple region by identifying the content in the region
Field corresponding with the region content.
8. a kind of device of image for identification, comprising:
Module is obtained, the acquisition module is configured as obtaining form image, and the form image includes in table line and table
Hold, the table line is associated with multiple fields in the table content;
Table line identification module, the table line identification module are configured as identifying the table line based on the form image;
Division module, the division module are configured as being divided into the form image multiple based on the table line identified
Region;And
Field identification module, the field identification module are configured as identifying described more based on the multiple region divided
The content of a field.
9. device according to claim 8, wherein the table line identification module includes:
Binarization block, the binarization block are configured as carrying out binaryzation to the form image to generate bianry image;
Hough transformation module, the Hough transformation module are configured as being directed to bianry image progress Hough transformation to generate
The hough space of the bianry image;
Extreme point determining module, the extreme point determining module are configured to determine that the extreme point in the hough space;And
Table line determining module, the table line determining module are configured as determining the table line based on the extreme point.
10. device according to claim 9, wherein the binarization block includes:
Noise reduction module, the noise reduction module are configured as carrying out noise reduction to the form image to generate the tabular drawing through noise reduction
Picture;And
First bianry image generation module, the first bianry image generation module are configured as to the tabular drawing through noise reduction
As carrying out binaryzation to generate bianry image.
11. device according to claim 9, wherein the binarization block includes:
Edge detection module, the edge detection module are configured as to the form image application edge detection to generate through side
The form image of edge detection;And
Second bianry image generation module, the second bianry image generation module are configured as to the table through edge detection
Table images carry out binaryzation to generate the bianry image.
12. device according to claim 9, wherein the extreme point determining module includes:
Threshold value obtains module, and the threshold value obtains module and is configured as obtaining predetermined threshold related with the feature of the extreme point
Value, the feature includes any one of number and size of the extreme point of the extreme point;And
Threshold value extreme point determining module, the threshold value extreme point determining module are configured as based on described in predetermined threshold determination
Extreme point.
13. device according to claim 8, wherein the division module includes:
Region division module, the region division module are configured as the seat based on the table line in the form image
Mark, is divided into the multiple region for the form image, and each region in the multiple region corresponds to the multiple word
Each field in section.
14. device according to claim 13, wherein the field identification module includes:
Region identification module, the region identification module are configured as passing through knowledge for a region in the multiple region
Content in the not described region determines the content of the field corresponding with the region in the multiple field.
15. a kind of equipment of image for identification, the equipment include:
One or more processors;And
Storage device, for storing one or more programs, when one or more of programs are by one or more of processing
Device executes, so that one or more of processors realize method according to any one of claims 1-7.
16. a kind of computer readable storage medium is stored thereon with computer program, realization when described program is executed by processor
Method according to any one of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810105171.9A CN110135218A (en) | 2018-02-02 | 2018-02-02 | The method, apparatus, equipment and computer storage medium of image for identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810105171.9A CN110135218A (en) | 2018-02-02 | 2018-02-02 | The method, apparatus, equipment and computer storage medium of image for identification |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110135218A true CN110135218A (en) | 2019-08-16 |
Family
ID=67567116
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810105171.9A Pending CN110135218A (en) | 2018-02-02 | 2018-02-02 | The method, apparatus, equipment and computer storage medium of image for identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110135218A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111695517A (en) * | 2020-06-12 | 2020-09-22 | 北京百度网讯科技有限公司 | Table extraction method and device for image, electronic equipment and storage medium |
CN111860315A (en) * | 2020-07-20 | 2020-10-30 | 中国建设银行股份有限公司 | Method, device and equipment for detecting form line and storage medium |
WO2023045277A1 (en) * | 2021-09-27 | 2023-03-30 | 上海合合信息科技股份有限公司 | Method and device for converting table in image into spreadsheet |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315668A (en) * | 2008-07-01 | 2008-12-03 | 上海大学 | Automatic detection method for test paper form |
CN101447017A (en) * | 2008-11-27 | 2009-06-03 | 浙江工业大学 | Method and system for quickly identifying and counting votes on the basis of layout analysis |
CN103258198A (en) * | 2013-04-26 | 2013-08-21 | 四川大学 | Extraction method for characters in form document image |
CN106897690A (en) * | 2017-02-22 | 2017-06-27 | 南京述酷信息技术有限公司 | PDF table extracting methods |
US9697423B1 (en) * | 2015-12-31 | 2017-07-04 | Konica Minolta Laboratory U.S.A., Inc. | Identifying the lines of a table |
-
2018
- 2018-02-02 CN CN201810105171.9A patent/CN110135218A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101315668A (en) * | 2008-07-01 | 2008-12-03 | 上海大学 | Automatic detection method for test paper form |
CN101447017A (en) * | 2008-11-27 | 2009-06-03 | 浙江工业大学 | Method and system for quickly identifying and counting votes on the basis of layout analysis |
CN103258198A (en) * | 2013-04-26 | 2013-08-21 | 四川大学 | Extraction method for characters in form document image |
US9697423B1 (en) * | 2015-12-31 | 2017-07-04 | Konica Minolta Laboratory U.S.A., Inc. | Identifying the lines of a table |
CN106897690A (en) * | 2017-02-22 | 2017-06-27 | 南京述酷信息技术有限公司 | PDF table extracting methods |
Non-Patent Citations (2)
Title |
---|
汪磊: "基于结构特征提取的选票分析系统的设计与研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
郭斯羽: "《面向检测的图像处理技术》", 31 August 2015, 湖南大学出版社 * |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111695517A (en) * | 2020-06-12 | 2020-09-22 | 北京百度网讯科技有限公司 | Table extraction method and device for image, electronic equipment and storage medium |
CN111695517B (en) * | 2020-06-12 | 2023-08-18 | 北京百度网讯科技有限公司 | Image form extraction method and device, electronic equipment and storage medium |
CN111860315A (en) * | 2020-07-20 | 2020-10-30 | 中国建设银行股份有限公司 | Method, device and equipment for detecting form line and storage medium |
WO2023045277A1 (en) * | 2021-09-27 | 2023-03-30 | 上海合合信息科技股份有限公司 | Method and device for converting table in image into spreadsheet |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11816165B2 (en) | Identification of fields in documents with neural networks without templates | |
EP3437019B1 (en) | Optical character recognition in structured documents | |
JP5418386B2 (en) | Image processing apparatus, image processing method, and program | |
CN111931664A (en) | Mixed note image processing method and device, computer equipment and storage medium | |
CN112016438A (en) | Method and system for identifying certificate based on graph neural network | |
US20140233837A1 (en) | Systems and methods for storing image properties for recreating an image | |
US20210064861A1 (en) | Identification of table partitions in documents with neural networks using global document context | |
US9031308B2 (en) | Systems and methods for recreating an image using white space and check element capture | |
CN110222641B (en) | Method and apparatus for recognizing image | |
JP2016048444A (en) | Document identification program, document identification device, document identification system, and document identification method | |
CN112862024B (en) | Text recognition method and system | |
CN110135218A (en) | The method, apparatus, equipment and computer storage medium of image for identification | |
CN110781890A (en) | Identification card identification method and device, electronic equipment and readable storage medium | |
CN111340022A (en) | Identity card information identification method and device, computer equipment and storage medium | |
CN113158895A (en) | Bill identification method and device, electronic equipment and storage medium | |
CN111160188A (en) | Financial bill identification method, device, equipment and storage medium | |
US11106908B2 (en) | Techniques to determine document recognition errors | |
US11881044B2 (en) | Method and apparatus for processing image, device and storage medium | |
CN113673528B (en) | Text processing method, text processing device, electronic equipment and readable storage medium | |
CN111209856A (en) | Invoice information identification method and device, electronic equipment and storage medium | |
EP4244761A1 (en) | Fraud detection via automated handwriting clustering | |
CN111414889B (en) | Financial statement identification method and device based on character identification | |
CN109493285A (en) | Image processing method, device, server and storage medium based on crowdsourcing | |
CN114299509A (en) | Method, device, equipment and medium for acquiring information | |
CN114495146A (en) | Image text detection method and device, computer 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 |