CN108133169A - Line processing method and device for text image - Google Patents
Line processing method and device for text image Download PDFInfo
- Publication number
- CN108133169A CN108133169A CN201611099659.2A CN201611099659A CN108133169A CN 108133169 A CN108133169 A CN 108133169A CN 201611099659 A CN201611099659 A CN 201611099659A CN 108133169 A CN108133169 A CN 108133169A
- Authority
- CN
- China
- Prior art keywords
- text
- connected domain
- image
- domain
- journey
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 20
- 238000000034 method Methods 0.000 claims abstract description 31
- 238000012545 processing Methods 0.000 claims description 25
- 238000004458 analytical method Methods 0.000 claims description 18
- 238000005286 illumination Methods 0.000 claims description 14
- 238000010586 diagram Methods 0.000 claims description 12
- 230000000694 effects Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000010276 construction Methods 0.000 claims description 6
- 238000003780 insertion Methods 0.000 claims description 6
- 230000037431 insertion Effects 0.000 claims description 6
- 238000001514 detection method Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000015654 memory Effects 0.000 description 15
- 230000006870 function Effects 0.000 description 11
- 238000004590 computer program Methods 0.000 description 8
- 238000012937 correction Methods 0.000 description 8
- 230000008569 process Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 6
- 230000004048 modification Effects 0.000 description 5
- 238000012986 modification Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 3
- 230000001154 acute effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004141 dimensional analysis Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000010295 mobile communication Methods 0.000 description 2
- 230000009182 swimming Effects 0.000 description 2
- 240000000233 Melia azedarach Species 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000008878 coupling Effects 0.000 description 1
- 238000010168 coupling process Methods 0.000 description 1
- 238000005859 coupling reaction Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- 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
- G06V30/413—Classification of content, e.g. text, photographs or tables
-
- 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
- G06V30/416—Extracting the logical structure, e.g. chapters, sections or page numbers; Identifying elements of the document, e.g. authors
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the application provides an in-line processing method and device for text images, wherein the method comprises the following steps: extracting an image connected domain of the text image, and classifying the image connected domain; initializing each text line by using the image connected domain and taking a preset width as a line window; and sequentially merging the image connected domains of all classes into all initialization text lines to which the image connected domains belong. The method and the device for analyzing the text image can analyze the layout of the text image accurately and conveniently and form a line.
Description
Technical field
The application belongs to image identification technical field, and in particular to it is a kind of for text image embark on journey processing method and its
Device.
Background technology
In the text image that image acquiring device is used to shoot, often due to the problem of shooting angle, can exist compared with
Big text distortion phenomenon causes to be difficult to effective printed page analysis to the text image of shooting, eventually lead to text diagram
As being difficult to identify.
Current printed page analysis system generally use text correction, the flow of printed page analysis.Text correction utilizes text
Domain information is connected, the text (tilting, perspective) of deformation is corrected, during embarking on journey, row is used to the text corrected
The mode of projection is completed to embark on journey.
But prior art needs first correct text, final result of the effect dependent on correction of embarking on journey.But
In shooting image process, since the angle of shooting is of all kinds, it is difficult to text image is carried out single correction (such as
In scan text image, inclined situation is generally only existed), it generally requires to carry out a variety of corrections.In addition, for formula text figure
Picture, since text information is limited, it is difficult to preferably be corrected using limited information, correction result is difficult to full in many cases
Foot projects rows of requirement.
Therefore, how the accurate space of a whole page for being easily directed to text image is analyzed and is embarked on journey, and is become in the prior art urgently
The technical issues of need to solving.
Invention content
One of the technical issues of the embodiment of the present application solves is to provide a kind of processing method of embarking on journey for text image
And its device, accurately easily it can be analyzed and be embarked on journey for the space of a whole page of text image.
The embodiment of the present application provides a kind of processing method of embarking on journey for text image, including：
The image connectivity domain of the text image is extracted, and is classified to described image connected domain；
Using described image connected domain, each line of text is initialized using preset width as window of embarking on journey；
Described image connected domain of all categories is incorporated to each initialization line of text of its ownership successively.
In the embodiment of the present application, the image connectivity domain of the extraction text image, and to described image connected domain
It is further included before being classified：
Remove the even effect of uneven illumination in the text image, and by the text image binaryzation.
In the embodiment of the present application, the even effect of uneven illumination in the removal text image, and by the text
It is further included after image binaryzation：
Picture size analysis is carried out to the text image after binary conversion treatment, obtains the text in the text image
Size.
In the embodiment of the present application, the image connectivity domain of the extraction text image, and to described image connected domain
Classification is carried out to include：
If the length and width of the connected domain are identical, the connected domain is classified as text connected domain；
If the area of the connected domain is more than or equal to preset area high threshold, the connected domain is classified as
Big connected domain；
If the area of the connected domain is less than preset area Low threshold, the connected domain is classified as small connection
Domain；
If the connected domain length-width ratio is more than or less than set high-low threshold value, the connected domain is classified as
Special connected domain.
In the embodiment of the present application, the image connectivity domain of the extraction text image, and to described image connected domain
Classification is carried out to further include：
The connected domain that size is less than to the text size of preset value is deleted as noise connected domain.
In the embodiment of the present application, the image connectivity domain of the extraction text image, and to described image connected domain
Classification is carried out to further include：
The width and height of the text connected domain boundary rectangle are counted, the two average value is calculated and obtains the text image
The width of middle text character is high.
In the embodiment of the present application, it is described using described image connected domain, it is initial using preset width as window of embarking on journey
Change each line of text to include：
The central point of text connected domain is detected whether in the center baseline range of the window of embarking on journey to be inserted entered a profession；
Such as in the range of, then the text connected domain is inserted into described to be inserted enter a profession, and update its row information；
If not in the range of, then newline is set up into the text connected domain position.
In the embodiment of the present application, each initialization that described image connected domain of all categories is incorporated to its ownership successively
Line of text includes：
Special connected domain, small connected domain and big connected domain are incorporated to successively in each initialization line of text.
In the embodiment of the present application, it is described that special connected domain, small connected domain and big connected domain are incorporated to each initialization successively
It is further included in line of text：
The high variation of row of the initialization line of text is enabled to be less than default changing value, and the high variation of the row is along the row
Center line symmetrically change.
In the embodiment of the present application, it further includes：
If there is overlapped initialization line of text, then it is new to merge the initialization line of text of the overlapping
Row.
The corresponding above method, the application provide a kind of processing unit of embarking on journey for text image, including：
Sort module for extracting the image connectivity domain of the text image, and classifies to described image connected domain；
For utilizing described image connected domain, each text is initialized using preset width as window of embarking on journey for initial module
Row；
Module is inserted into, for described image connected domain of all categories to be incorporated to each initialization line of text of its ownership successively.
In the embodiment of the present application, it further includes：
Preprocessing module, for removing the even effect of uneven illumination in the text image, and by the text image two
Value.
In the embodiment of the present application, it further includes：
Analysis module for carrying out picture size analysis to the text image after binary conversion treatment, obtains the text
Text size in this image.
In the embodiment of the present application, the sort module includes：
The connected domain if the length and width for the connected domain are identical, are classified as text and connected by text classification unit
Logical domain；
Macrotaxonomy unit, will if the area for the connected domain is more than or equal to preset area high threshold
The connected domain is classified as big connected domain；
Subclassification unit, if the area for the connected domain is less than preset area Low threshold, by the connection
Domain is classified as small connected domain；
Special sort unit, if being more than or less than set high-low threshold value for the connected domain length-width ratio,
The connected domain is classified as special connected domain.
In the embodiment of the present application, the sort module further includes：
Filter element, the connected domain for size to be less than to the text size of preset value are deleted as noise connected domain
It removes.
In the embodiment of the present application, the sort module further includes：
Dimension cells for counting the width and height of the text connected domain boundary rectangle, calculate the two average value and obtain
The width for obtaining text character in the text image is high.
In the embodiment of the present application, the initial module includes：
Detection unit, for detect the central point of text connected domain whether the window of embarking on journey to be inserted entered a profession center baseline
In the range of；
The text connected domain for such as in the range of, is then inserted into described to be inserted enter a profession, and update by first Construction Bank's unit
Its row information；
Second Construction Bank's unit, for if not in the range of, then the text connected domain position being set up newline.
In the embodiment of the present application, the insertion module is specifically used for special connected domain, small connected domain and big connected domain
It is incorporated to successively in each initialization line of text.
In the embodiment of the present application, the variation that the insertion module is additionally operable to enable the row of the initialization line of text high is less than
Default changing value, and the high variation of the row symmetrically changes along the center line of the row.
In the embodiment of the present application, it further includes：
Merging module, for if there is overlapped initialization line of text, then by the initialization text of the overlapping
Row carries out merging into newline.
The embodiment of the present application extracts the image connectivity domain of the text image, and classifies to described image connected domain,
Using described image connected domain, each line of text is initialized using preset width as window of embarking on journey, described in will be of all categories
Image connectivity domain is incorporated to each initialization line of text of its ownership successively.Therefore, in printed page analysis process, the embodiment of the present application
It does not need to carry out any correction to text, can directly be embarked on journey process by the information of connected domain, and be directed to all kinds of figures
As connected domain has carried out different processing, make to all kinds of well adapting to property of image.
Description of the drawings
In order to illustrate the technical solutions in the embodiments of the present application or in the prior art more clearly, to embodiment or it will show below
There is attached drawing needed in technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments described in application, for those of ordinary skill in the art, can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is a kind of one embodiment flow chart of processing method of embarking on journey for text image that the application provides；
Fig. 2 is an embodiment flow of step S1 in a kind of processing method of embarking on journey for text image that the application provides
Figure；
Fig. 3 is that the application connected domain carries out one exemplary schematic diagram of classification；
Fig. 4 is another embodiment stream of step S1 in a kind of processing method of embarking on journey for text image that the application provides
Cheng Tu；
Fig. 5 is one exemplary schematic diagram of the application erased noise connected domain；
Fig. 6 is the another embodiment stream of step S1 in a kind of processing method of embarking on journey for text image that the application provides
Cheng Tu；
Fig. 7 is an embodiment flow of step S2 in a kind of processing method of embarking on journey for text image that the application provides
Figure；
Fig. 8 is a kind of another embodiment flow chart of processing method of embarking on journey for text image that the application provides；
Fig. 9 a to Fig. 9 c are an exemplary schematic diagrames of the application step S01 processing text images；
Figure 10 is a kind of processing method another embodiment flow chart of embarking on journey for text image that the application provides；
Figure 11 is a kind of processing method another embodiment flow chart of embarking on journey for text image that the application provides；
Figure 12 is an exemplary schematic diagram of the application step S4 processing text images；
Figure 13 is a kind of one example structure figure of processing unit of embarking on journey for text image that the application provides；
Figure 14 is an embodiment of sort module in a kind of processing unit of embarking on journey for text image that the application provides
Structure chart；
Figure 15 is another implementation of sort module in a kind of processing unit of embarking on journey for text image that the application provides
Example structure chart；
Figure 16 is another implementation of sort module in a kind of processing unit of embarking on journey for text image that the application provides
Example structure chart；
Figure 17 is an embodiment of initial module in a kind of processing unit of embarking on journey for text image that the application provides
Structure chart；
Figure 18 is another example structure figure of a kind of processing unit of embarking on journey for text image that the application provides；
Figure 19 is a kind of processing unit another embodiment structure chart of embarking on journey for text image that the application provides；
Figure 20 is a kind of processing unit another embodiment structure chart of embarking on journey for text image that the application provides；
Figure 21 is the hardware configuration signal of the electronic equipment for the processing method of embarking on journey for text image that the application provides
Figure.
Specific embodiment
The embodiment of the present application extracts the image connectivity domain of the text image, and classifies to described image connected domain,
Using described image connected domain, each line of text is initialized using preset width as window of embarking on journey, described in will be of all categories
Image connectivity domain is incorporated to each initialization line of text of its ownership successively.Therefore, in printed page analysis process, the embodiment of the present application
It does not need to carry out any correction to text, can directly be embarked on journey process by the information of connected domain, and be directed to all kinds of figures
As connected domain has carried out different processing, make to all kinds of well adapting to property of image.
Although the application can have many various forms of embodiments, in the accompanying drawings display and will herein in detail
The specific embodiment of description, it should be appreciated that the disclosure of this embodiment should be considered as the example of principle, and be not intended to this Shen
It please be limited to the specific embodiment being shown and described.In the following description, identical label shows for describing the several of attached drawing
Identical, similar or corresponding part in figure.
As used herein, "one" or " one kind " of term are defined as one (kind) or more than one (kind).As herein
It is used, term " multiple " is defined as two or more than two.As used herein, term " other " is defined as at least again
It is one or more.As used herein, term "comprising" and/or " having " are defined to include (that is, open language).Such as
Used herein, term " coupling " is defined as connecting, but is not necessarily to be directly connected to, and is not necessarily mechanically to connect.
As used herein, term " program " or " computer program " or similar terms are defined as designed on the computer systems
The instruction sequence of execution." program " or " computer program " may include subprogram, function, process, object method, object implementatio8,
Executable application, applet, servlet, source code, object code, shared library/dynamic load library and/or design are used
In the other instruction sequences performed on the computer systems.
Table is referred to " one embodiment ", " some embodiments ", " embodiment " or similar terms in entire this document
Show that a particular feature, structure, or characteristic described in conjunction with the embodiments is included at least one embodiment of the invention.Therefore, exist
The appearance of this word in the various places of entire this specification need not all represent identical embodiment.It is in addition, described specific
Feature, structure or characteristic can combine in any suitable manner in one or more embodiments without limitation.
As used herein, term "or" should be construed as inclusive or represent any one or any group
It closes.Therefore, " A, B or C " expression " following any one：A；B；C；A and B；A and C；B and C；A, B and C ".Only when element,
When function, step or the combination of action inherently mutually exclusive in some way, it will the exception of this definition occurs.
In order to which those skilled in the art is made to more fully understand the technical solution in the application, below in conjunction with the embodiment of the present application
In attached drawing, the technical solution in the embodiment of the present application is clearly and completely described, it is clear that described embodiment is only
It is some embodiments of the present application, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's all other embodiments obtained should all belong to the range of the application protection.
Further illustrate that the application implements with reference to illustrations.
Referring to Fig. 1, one embodiment of the application provides a kind of processing method of embarking on journey for text image, including：
The image connectivity domain of S1, the extraction text image, and classify to described image connected domain.
Visually apparently, the point to communicate with each other forms a region, and disconnected point forms different regions.
Such a all points communicate with each other the set formed, we are known as a connected region.
In the application one in the specific implementation, referring to Fig. 2, the step S1 includes：
If S11, the length and width of the connected domain are identical, the connected domain is classified as text connected domain.
If the area of S12, the connected domain is more than or equal to preset area high threshold, by the connected domain point
Class is big connected domain.
Specifically, the area high threshold is area>400*width*width, wherein width are text in text image
The width of the stroke of character.
If the area of S13, the connected domain is less than preset area Low threshold, the connected domain is classified as little Lian
Logical domain.
If S14, the connected domain length-width ratio are more than or less than set high-low threshold value, by the connected domain point
Class is special connected domain.
Specifically, the special connected domain includes：The symbols such as branch in formula, braces.
Referring to Fig. 3, label 31 is text connected domain in the text image 3 of acquisition, and big connection is not included in text image 3
Domain, label 32 are small connected domain, and label 33 is special connected domain.
It is another in the application in the specific implementation, referring to Fig. 4, the step S1 is further included：
S10, connected domain of the size less than the text size of preset value is deleted as noise connected domain.
Before the application classifies to image connectivity domain, first using size be less than preset value text size connected domain as
Noise connected domain is deleted.That is, if the size of connected domain is less than the text size of preset value, assert that the connected domain is
Noise connected domain, is deleted.
Specifically, the connected domain after 3 erased noise connected domain of text image is as shown in Figure 5.
Another in the specific implementation, referring to Fig. 6 in the application, the step S1 is further included：
The width and height of S15, the statistics text connected domain boundary rectangle calculate the two average value and obtain the text
The width of text character is high in image.
Specifically, the application calculates all boundary rectangle width according to the width and height of boundary rectangle in Fig. 4 and Fig. 5
Average value so as to obtain the width of text character in the text image, calculates all boundary rectangle height averages, so as to obtain
Obtain the height of text character in the text image.
S2, using described image connected domain, initialize each line of text using preset width as window of embarking on journey.
Specifically, the preset width is 4 times of character durations, and those of ordinary skill in the art can also use according to demand
Other preset values.
Another in the specific implementation, referring to Fig. 7 in the application, the step S2 includes：
S21, the central point of text connected domain is detected whether in the center baseline range of the window of embarking on journey to be inserted entered a profession.
Specifically, the application first from left to right sorts the text connected domain according to x coordinate, then will be after sequence
The text connected domain is embarked on journey successively according to its x coordinate and y-coordinate.
During the embarking on journey of the text connected domain, in order to reduce since line of text distortion caused by shooting is made to embarking on journey
Into influence, choose 4 times of width as window of embarking on journey.In the window, it can be assumed that there is no distortion changes for text connected domain
Shape, window of embarking on journey slide change with the addition of text connected domain.Criterion of embarking on journey is：When text connected domain is in window of embarking on journey
When being less than default center baseline range with the connected domain center parallax range entered a profession, the connected domain belongs to the row, by described in
Connected domain is entered a profession, and otherwise, the connected domain is set up newline.
S22, such as in the range of, then the text connected domain is inserted into described to be inserted enter a profession, and update its row information.
S23, if not in the range of, then newline is set up into the text connected domain position.
Specifically, during the application judges whether that all text connected domains are already inserted into and are embarked on journey, if also presence is not inserted
Enter the text connected domain in being embarked on journey, then return to step S21；If there is no the text connected domain being not inserted into being embarked on journey, then
Terminate to initialize each line of text.
S3, each initialization line of text that described image connected domain of all categories is incorporated to its ownership successively.
Due to initializing each line of text using the text connected domain, the step S3 is specially：
Special connected domain, small connected domain and big connected domain are incorporated to successively in each initialization line of text.
After row information initialization is completed, special connected domain, small connected domain and big connected domain are incorporated to each row successively successively
In the middle, the rule being incorporated to is：After being incorporated to the connected domain, there is not acute variation in row height, when going out existing High variation, it is necessary to assure
Capable upper edge and lower edge changes simultaneously, and changes basically identical.
The step S3 is further included：
The high variation of row of the initialization line of text is enabled to be less than default changing value, and the high variation of the row is along the row
Center line symmetrically change.Therefore, the application ensures that special connected domain, small connected domain and big connected domain are incorporated to each initialization successively
In line of text, row height symmetrically changes along the center line of the row, and changes and be less than default changing value.
In another embodiment of the application, including above-mentioned steps S1-S3, referring to Fig. 8, further included before the step S1：
The even effect of uneven illumination in S01, the removal text image, and by the text image binaryzation.
Specifically, due to the limitation of shooting condition, the text image of shooting can have that uneven illumination is even, pretreatment
Process needs to remove the even effect of uneven illumination, and carry out binaryzation to text image.
Referring specifically to Fig. 9 a to Fig. 9 c, Fig. 9 a are the gray document image of input, and Fig. 9 b are that removal background illumination influences
Text image, Fig. 9 c are the image of binaryzation.
In another embodiment of the application, including above-mentioned steps S1-S3, referring to Figure 10, also wrapped after the step S01
It includes：
S02, picture size analysis is carried out to the text image after binary conversion treatment, obtained in the text image
Text size.
Since text image comes from image capture device shooting, in many cases, due to shooting angle, the limit of distance
It makes, the size neither one fixed range of text in image.In order to unified analysis scale, need it is existing to image text into
Row dimensional analysis.
Specifically, the analysis of the application image text scale calculates text pen using the black run length in statistical picture
The strategy of width is drawn, i.e., the highest black distance of swimming of the frequency of occurrences is the stroke width of text as in the text image in the picture
Text size.
In another embodiment of the application, including above-mentioned steps S1-S3, referring to Figure 11, the method further includes：
S4, if there is overlapped initialization line of text, then the initialization line of text of the overlapping is merged
For newline.
It, will be described heavy when there are (Figure 12 a) during overlapped initialization line of text referring specifically to attached drawing 12a and Figure 12 b
Folded initialization line of text is merged as newline (Figure 12 b).
Referring to Figure 13, the corresponding above method, another embodiment of the application provides a kind of processing of embarking on journey for text image
Device, including：
Sort module 131 for extracting the image connectivity domain of the text image, and divides described image connected domain
Class.
For utilizing described image connected domain, each text is initialized using preset width as window of embarking on journey for initial module 132
One's own profession.
Module 133 is inserted into, for described image connected domain of all categories to be incorporated to each initialization text of its ownership successively
Row.
Visually apparently, the point to communicate with each other forms a region, and disconnected point forms different regions.
Such a all points communicate with each other the set formed, we are known as a connected region.
In the application one in the specific implementation, referring to Figure 14, the sort module 131 includes：
The connected domain if the length and width for the connected domain are identical, is classified as text by text classification unit 1311
This connected domain；
Macrotaxonomy unit 1312, if the area of text stroke for being included in the connected domain is more than or equal in advance
If area high threshold, then the connected domain is classified as big connected domain；
Subclassification unit 1313, if the area of text stroke for being included in the connected domain is less than preset area
The connected domain is then classified as small connected domain by Low threshold；
Special sort unit 1314, if being more than or less than set height threshold for the connected domain length-width ratio
Value, then be classified as special connected domain by the connected domain.
Specifically, the area high threshold is area>400*width*width, wherein width are text in text image
The width of the stroke of character.
Specifically, the special connected domain includes：The symbols such as branch in formula, braces.
Referring to Fig. 3, label 31 is text connected domain in the text image 3 of acquisition, and big connection is not included in text image 3
Domain, label 32 are small connected domain, and label 33 is special connected domain.
It is another in the application in the specific implementation, referring to Figure 15, the sort module 131 further includes：
Filter element 1315 connects for deleting size using the text size less than the noise of preset value text size
Domain.
Before the application classifies to image connectivity domain, first using size be less than preset value text size connected domain as
Noise connected domain is deleted.That is, if the size of connected domain is less than the text size of preset value, assert that the connected domain is
Noise connected domain, is deleted.
Specifically, the connected domain after 3 erased noise connected domain of text image is as shown in Figure 5.
Another in the specific implementation, referring to Figure 16 in the application, the sort module 131 further includes：
For counting the width and height of the text connected domain boundary rectangle, it is average to calculate the two for dimension cells 1316
The width that value obtains text character in the text image is high.
Specifically, the application calculates all boundary rectangle width according to the width and width of boundary rectangle in Fig. 4 and Fig. 5
Average value so as to obtain the height of text character in the text image, calculates all boundary rectangle height averages, so as to obtain
Obtain the length of text character in the text image.
Specifically, for the preset length of window of embarking on journey for 4 times of character durations, those of ordinary skill in the art also can basis
Demand uses other preset values.
Another in the specific implementation, referring to Figure 17 in the application, the initial module 122 includes：
Detection unit 1221, for whether detecting the central point of text connected domain at the center of window of embarking on journey to be inserted entered a profession
In baseline range.
First Construction Bank's unit 1222, for such as in the range of, then by the text connected domain be inserted into it is described it is to be inserted enter a profession, and
Update its row information.
Second Construction Bank's unit 1223, for if not in the range of, then the text connected domain position being set up newline.
Specifically, the application first from left to right sorts the text connected domain according to x coordinate, then will be after sequence
The text connected domain is embarked on journey successively according to its x coordinate and y-coordinate.
During the embarking on journey of the text connected domain, in order to reduce the influence distorted due to line of text caused by shooting,
4 times of character durations are chosen as window of embarking on journey.In the window, it can be assumed that torsional deformation is not present in text connected domain row,
Window of embarking on journey slides change with the addition of text connected domain.Criterion of embarking on journey is：When text connected domain in window of embarking on journey with
When the connected domain center parallax range entered a profession is less than default center baseline range, the connected domain belongs to the row, by the company
Logical domain is entered a profession, and otherwise, the connected domain is set up newline.
Specifically, during the application judges whether that all text connected domains are already inserted into and are embarked on journey, if also presence is not inserted
Enter the text connected domain in being embarked on journey, then return to detection unit；If there is no the text connected domain being not inserted into being embarked on journey, then
Terminate to initialize each line of text.
Due to initializing each line of text using the text connected domain, the insertion module 123 is specifically used for special company
Logical domain, small connected domain and big connected domain are incorporated to successively in each initialization line of text.
After row information initialization is completed, special connected domain, small connected domain and big connected domain are incorporated to each row successively successively
In the middle, the rule being incorporated to is：After being incorporated to the connected domain, there is not acute variation in row height, when going out existing High variation, it is necessary to assure
Capable upper edge and lower edge changes simultaneously, and changes basically identical.
The step is inserted into the variation that module 123 is additionally operable to enable the row of the initialization line of text high and is less than default variation
Value, and the high variation of the row symmetrically changes along the center line of the row.Therefore, the application ensures special connected domain, small connection
Domain and big connected domain are incorporated to successively in each initialization line of text, and row height symmetrically changes along the center line of the row, and variation is less than
Default changing value.
In another embodiment of the application, including above-mentioned sort module 131, initial module 132, module 133 is inserted into, referring to
Figure 18 is further included：
Preprocessing module 130, for removing the even effect of uneven illumination in the text image, and by the text image
Binaryzation.
Specifically, due to the limitation of shooting condition, the text image of shooting can have that uneven illumination is even, pretreatment
Process needs to remove the even effect of uneven illumination, and carry out binaryzation to text image.
Referring specifically to Fig. 9 a to Fig. 9 c, Fig. 9 a are the gray document image of input, and Fig. 9 b are that removal background illumination influences
Text image, Fig. 9 c are the image of binaryzation.
In another embodiment of the application, including above-mentioned sort module 131, initial module 132, module 133 is inserted into, referring to
Figure 19 is further included：
Analysis module 1300 for carrying out picture size analysis to the text image after binary conversion treatment, obtains institute
State the text size in text image.
Since text image comes from image capture device shooting, in many cases, due to shooting angle, the limit of distance
It makes, the size neither one fixed range of text in image.In order to unified analysis scale, need it is existing to image text into
Row dimensional analysis.
Specifically, the analysis of the application image text scale calculates text pen using the black run length in statistical picture
The strategy of width is drawn, i.e., the highest black distance of swimming of the frequency of occurrences is the stroke width of text as in the text image in the picture
Text size.
In another embodiment of the application, including above-mentioned sort module 131, initial module 132, module 133 is inserted into, referring to
Figure 20 is further included：
Merging module 134, for if there is overlapped initialization line of text, then by the initialization of overlapping text
One's own profession is merged as newline.
Referring specifically to attached drawing 11, when there are during overlapped initialization line of text, by the initialization text of the overlapping
Row carries out merging into newline.
Figure 21 is hardware architecture diagram of the application for the electronic equipment of the processing method of embarking on journey of text image.According to
Shown in Figure 21, which includes：
One or more processors 2110 and memory 2120, in Figure 21 by taking a processor 2110 as an example.
It can also include for the equipment of the processing method of embarking on journey of text image：Input unit 2130 and output device
2130。
Processor 2110, memory 2120, input unit 2130 and output device 2130 can by bus or other
Mode connects, in Figure 21 for being connected by bus.
Memory 2120 is used as a kind of non-volatile computer readable storage medium storing program for executing, available for storing non-volatile software journey
Sequence, non-volatile computer executable program and module, such as the processing of embarking on journey for text image in the embodiment of the present application
Corresponding program instruction/the module of method is (for example, sort module 131, initial module 132, insertion module shown in attached drawing 13
133).Processor 2110 is stored in non-volatile software program, instruction and module in memory 2120 by operation, so as to
The various function application of execute server and data processing realize that above method embodiment is directed to the place that embarks on journey of text image
Reason method.
Memory 2120 can include storing program area and storage data field, wherein, storing program area can store operation system
System, the required application program of at least one function；Storage data field can be stored according to the processing dress of embarking on journey for text image
That puts uses created data etc..In addition, memory 2120 can include high-speed random access memory 2120, can also wrap
Include nonvolatile memory 2120, for example, at least a magnetic disk storage 2120, flush memory device or other nonvolatile solid states
Memory 2120.In some embodiments, memory 2120 is optional including relative to the remotely located storage of processor 2110
Device 2120, these remote memories 2120 can pass through network connection to audio mode selector.The example packet of above-mentioned network
It includes but is not limited to internet, intranet, LAN, mobile radio communication and combinations thereof.
Input unit 2130 can receive the place that embarks on journey of the number or character information and generation of input with being directed to text image
Manage the user setting of device and the key signals input that function control is related.Output device 2130 may include the equipment such as loud speaker.
One or more of modules are stored in the memory 2120, when by one or more of processors
During 2110 execution, the processing method of embarking on journey for text image in above-mentioned any means embodiment is performed.
The said goods can perform the method that the embodiment of the present application is provided, and has the corresponding function module of execution method and has
Beneficial effect.The not technical detail of detailed description in the present embodiment, reference can be made to the method that the embodiment of the present application is provided.
The electronic equipment of the embodiment of the present application exists in a variety of forms, including but not limited to:
(1) mobile communication equipment:The characteristics of this kind equipment is that have mobile communication function, and to provide speech, data
It communicates as main target.This Terminal Type includes:Smart mobile phone (such as iPhone), multimedia handset, functional mobile phone and low
Hold mobile phone etc..
(2) super mobile personal computer equipment:This kind equipment belongs to the scope of personal computer, there is calculating and processing work(
Can, generally also have mobile Internet access characteristic.This Terminal Type includes:PDA, MID and UMPC equipment etc., such as iPad.
(3) portable entertainment device:This kind equipment can show and play multimedia content.The kind equipment includes:Audio,
Video player (such as iPod), handheld device, e-book and intelligent toy and portable car-mounted navigation equipment.
(4) server:The equipment for providing the service of calculating, the composition of server are total including processor, hard disk, memory, system
Line etc., server is similar with general computer architecture, but due to needing to provide highly reliable service, in processing energy
Power, stability, reliability, safety, scalability, manageability etc. are more demanding.
(21) other have the function of the electronic device of data interaction.
The apparatus embodiments described above are merely exemplary, wherein the module illustrated as separating component can
To be or may not be physically separate, the component shown as module may or may not be physics mould
Block, you can be located at a place or can also be distributed on multiple network modules.It can be selected according to the actual needs
In some or all of module realize the purpose of this embodiment scheme.Those of ordinary skill in the art are not paying creativeness
Labour in the case of, you can to understand and implement.
It will be understood by those skilled in the art that embodiments herein can be provided as method, apparatus (equipment) or computer
Program product.Therefore, in terms of the application can be used complete hardware embodiment, complete software embodiment or combine software and hardware
Embodiment form.Moreover, the meter for wherein including computer usable program code in one or more can be used in the application
The computer journey that calculation machine usable storage medium is implemented on (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of sequence product.
The application is with reference to the method, apparatus (equipment) of embodiment and the flow chart and/or box of computer program product
Figure describes.It should be understood that each flow and/or the side in flowchart and/or the block diagram can be realized by computer program instructions
The combination of flow and/or box in frame and flowchart and/or the block diagram.These computer program instructions can be provided to logical
With the processor of computer, special purpose computer, Embedded Processor or other programmable data processing devices to generate a machine
Device so that the instruction generation performed by computer or the processor of other programmable data processing devices is used to implement in flow
The device of function specified in one flow of figure or multiple flows and/or one box of block diagram or multiple boxes.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works so that the instruction generation being stored in the computer-readable memory includes referring to
Enable the manufacture of device, the command device realize in one flow of flow chart or multiple flows and/or one box of block diagram or
The function of being specified in multiple boxes.
These computer program instructions can be also loaded into computer or other programmable data processing devices so that counted
Series of operation steps are performed on calculation machine or other programmable devices to generate computer implemented processing, so as in computer or
The instruction offer performed on other programmable devices is used to implement in one flow of flow chart or multiple flows and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Although the preferred embodiment of the application has been described, those skilled in the art once know basic creation
Property concept, then additional changes and modifications may be made to these embodiments.So appended claims be intended to be construed to include it is excellent
It selects embodiment and falls into all change and modification of the application range.Obviously, those skilled in the art can be to the application
Various modification and variations are carried out without departing from spirit and scope.If in this way, these modifications and variations of the application
Belong within the scope of the application claim and its equivalent technologies, then the application is also intended to exist comprising these modification and variations
It is interior.
Claims (20)
1. a kind of processing method of embarking on journey for text image, which is characterized in that including：
The image connectivity domain of the text image is extracted, and is classified to described image connected domain；
Using described image connected domain, each line of text is initialized using preset width as window of embarking on journey；
Described image connected domain of all categories is incorporated to each initialization line of text of its ownership successively.
2. the method as described in claim 1, which is characterized in that the image connectivity domain of the extraction text image, and it is right
Described image connected domain further includes before being classified：
Remove the even effect of uneven illumination in the text image, and by the text image binaryzation.
3. method as claimed in claim 2, which is characterized in that the even effect of uneven illumination in the removal text image
Should, and will be further included after the text image binaryzation：
Picture size analysis is carried out to the text image after binary conversion treatment, obtains the text ruler in the text image
It is very little.
4. method as described in claim 1, which is characterized in that the image connectivity domain of the extraction text image, and to institute
It states image connectivity domain and classify and include：
If the length and width of the connected domain are identical, the connected domain is classified as text connected domain；
If the area of the connected domain is more than or equal to preset area high threshold, the connected domain is classified as Dalian
Logical domain；
If the area of the connected domain is less than preset area Low threshold, the connected domain is classified as small connected domain；
If the connected domain length-width ratio is more than or less than set high-low threshold value, it is special that the connected domain is classified as
Connected domain.
5. method as claimed in claim 4, which is characterized in that the image connectivity domain of the extraction text image, and it is right
Described image connected domain carries out classification and further includes：
The connected domain that size is less than to the text size of preset value is deleted as noise connected domain.
6. method as claimed in claim 5, which is characterized in that the image connectivity domain of the extraction text image, and it is right
Described image connected domain carries out classification and further includes：
The width and height of the text connected domain boundary rectangle are counted, the two average value is calculated and obtains the text image Chinese
The width of this character is high.
7. method as claimed in claim 4, which is characterized in that it is described to utilize described image connected domain, made with preset width
Each line of text is initialized for window of embarking on journey to include：
The central point of text connected domain is detected whether in the center baseline range of the window of embarking on journey to be inserted entered a profession；
Such as in the range of, then the text connected domain is inserted into described to be inserted enter a profession, and update its row information；
If not in the range of, then newline is set up into the text connected domain position.
8. the method for claim 7, which is characterized in that described that described image connected domain of all categories is incorporated to it successively
Each initialization line of text of ownership includes：
Special connected domain, small connected domain and big connected domain are incorporated to successively in each initialization line of text.
9. method as claimed in claim 8, which is characterized in that it is described by special connected domain, small connected domain and big connected domain according to
Secondary be incorporated in each initialization line of text further includes：
The high variation of row of the initialization line of text is enabled to be less than default changing value, and the high variation of the row is along the row
Heart line symmetrically changes.
10. the method as described in claim 1, which is characterized in that further include：
If there is overlapped initialization line of text, then the initialization line of text of the overlapping is merged as newline.
11. a kind of processing unit of embarking on journey for text image, which is characterized in that including：
Sort module for extracting the image connectivity domain of the text image, and classifies to described image connected domain；
For utilizing described image connected domain, each line of text is initialized using preset width as window of embarking on journey for initial module；
Module is inserted into, for described image connected domain of all categories to be incorporated to each initialization line of text of its ownership successively.
12. device as claimed in claim 11, which is characterized in that further include：
Preprocessing module, for removing the even effect of uneven illumination in the text image, and by the text image binaryzation.
13. device as claimed in claim 12, which is characterized in that further include：
Analysis module for carrying out picture size analysis to the text image after binary conversion treatment, obtains the text diagram
Text size as in.
14. device as claimed in claim 11, which is characterized in that the sort module includes：
The connected domain if the length and width for the connected domain are identical, is classified as text connected domain by text classification unit；
Macrotaxonomy unit, if the area for the connected domain is more than or equal to preset area high threshold, by described in
Connected domain is classified as big connected domain；
Subclassification unit, if the area for the connected domain is less than preset area Low threshold, by the connected domain point
Class is small connected domain；
Special sort unit, if being more than or less than set high-low threshold value for the connected domain length-width ratio, by institute
It states connected domain and is classified as special connected domain.
15. device as claimed in claim 14, which is characterized in that the sort module further includes：
Filter element, the connected domain for size to be less than to the text size of preset value are deleted as noise connected domain.
16. device as claimed in claim 15, which is characterized in that the sort module further includes：
Dimension cells for counting the width and height of the text connected domain boundary rectangle, calculate the two average value and obtain institute
The width for stating text character in text image is high.
17. device as claimed in claim 14, which is characterized in that the initial module includes：
Detection unit, for detect the central point of text connected domain whether the window of embarking on journey to be inserted entered a profession center baseline range
It is interior；
The text connected domain for such as in the range of, is then inserted into described to be inserted enter a profession, and update its row by first Construction Bank's unit
Information；
Second Construction Bank's unit, for if not in the range of, then the text connected domain position being set up newline.
18. device as claimed in claim 17, which is characterized in that the insertion module is specifically used for special connected domain, small
Connected domain and big connected domain are incorporated to successively in each initialization line of text.
19. device as claimed in claim 18, which is characterized in that the insertion module is additionally operable to enable the initialization line of text
The high variation of row be less than default changing value, and the high variation of the row symmetrically changes along the center line of the row.
20. device as claimed in claim 10, which is characterized in that further include：
Merging module, for if there is overlapped initialization line of text, then the initialization text of the overlapping being advanced
Row merges into newline.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611099659.2A CN108133169A (en) | 2016-12-01 | 2016-12-01 | Line processing method and device for text image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611099659.2A CN108133169A (en) | 2016-12-01 | 2016-12-01 | Line processing method and device for text image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108133169A true CN108133169A (en) | 2018-06-08 |
Family
ID=62388265
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611099659.2A Pending CN108133169A (en) | 2016-12-01 | 2016-12-01 | Line processing method and device for text image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108133169A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508244A (en) * | 2018-10-18 | 2019-03-22 | 北京新唐思创教育科技有限公司 | Data processing method and computer readable medium |
| CN109871743A (en) * | 2018-12-29 | 2019-06-11 | 口碑(上海)信息技术有限公司 | The localization method and device of text data, storage medium, terminal |
| CN109919155A (en) * | 2019-03-13 | 2019-06-21 | 厦门商集网络科技有限责任公司 | A kind of the inclination angle antidote and terminal of text image |
| CN109948598A (en) * | 2019-05-15 | 2019-06-28 | 达而观信息科技（上海）有限公司 | Document layout intelligent analysis method and device |
| WO2020177584A1 (en) * | 2019-03-01 | 2020-09-10 | 华为技术有限公司 | Graphic typesetting method and related device |
| CN114067150A (en) * | 2021-11-19 | 2022-02-18 | 王海丰 | Method and system for generating text line in any direction |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050271275A1 (en) * | 2004-06-04 | 2005-12-08 | Benq Corporation | Text character identification system and method thereof |
| CN102314608A (en) * | 2010-06-30 | 2012-01-11 | 汉王科技股份有限公司 | Method and device for extracting rows from character image |
| CN102567300A (en) * | 2011-12-29 | 2012-07-11 | 方正国际软件有限公司 | Picture document processing method and device |
| CN104182744A (en) * | 2013-05-24 | 2014-12-03 | 佳能株式会社 | Text detection method and device, and text message extraction method and system |
| CN105844207A (en) * | 2015-01-15 | 2016-08-10 | 富士通株式会社 | Text line extraction method and text line extraction equipment |
| CN106156772A (en) * | 2015-03-25 | 2016-11-23 | 佳能株式会社 | For determining the method and apparatus of word spacing and for the method and system of participle |
-
2016
- 2016-12-01 CN CN201611099659.2A patent/CN108133169A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050271275A1 (en) * | 2004-06-04 | 2005-12-08 | Benq Corporation | Text character identification system and method thereof |
| CN102314608A (en) * | 2010-06-30 | 2012-01-11 | 汉王科技股份有限公司 | Method and device for extracting rows from character image |
| CN102567300A (en) * | 2011-12-29 | 2012-07-11 | 方正国际软件有限公司 | Picture document processing method and device |
| CN104182744A (en) * | 2013-05-24 | 2014-12-03 | 佳能株式会社 | Text detection method and device, and text message extraction method and system |
| CN105844207A (en) * | 2015-01-15 | 2016-08-10 | 富士通株式会社 | Text line extraction method and text line extraction equipment |
| CN106156772A (en) * | 2015-03-25 | 2016-11-23 | 佳能株式会社 | For determining the method and apparatus of word spacing and for the method and system of participle |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508244A (en) * | 2018-10-18 | 2019-03-22 | 北京新唐思创教育科技有限公司 | Data processing method and computer readable medium |
| CN109508244B (en) * | 2018-10-18 | 2021-03-12 | 北京新唐思创教育科技有限公司 | Data processing method and computer readable medium |
| CN109871743A (en) * | 2018-12-29 | 2019-06-11 | 口碑(上海)信息技术有限公司 | The localization method and device of text data, storage medium, terminal |
| WO2020177584A1 (en) * | 2019-03-01 | 2020-09-10 | 华为技术有限公司 | Graphic typesetting method and related device |
| US11790584B2 (en) | 2019-03-01 | 2023-10-17 | Huawei Technologies Co., Ltd. | Image and text typesetting method and related apparatus thereof |
| CN109919155A (en) * | 2019-03-13 | 2019-06-21 | 厦门商集网络科技有限责任公司 | A kind of the inclination angle antidote and terminal of text image |
| CN109948598A (en) * | 2019-05-15 | 2019-06-28 | 达而观信息科技（上海）有限公司 | Document layout intelligent analysis method and device |
| CN114067150A (en) * | 2021-11-19 | 2022-02-18 | 王海丰 | Method and system for generating text line in any direction |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108133169A (en) | Line processing method and device for text image | |
| CN109325954B (en) | Image segmentation method and device and electronic equipment | |
| CN109255352B (en) | Target detection method, device and system | |
| CN109697416B (en) | Video data processing method and related device | |
| US8792722B2 (en) | Hand gesture detection | |
| US8750573B2 (en) | Hand gesture detection | |
| CN110363091B (en) | Face recognition method, device and equipment under side face condition and storage medium | |
| CN110222573B (en) | Face recognition method, device, computer equipment and storage medium | |
| CN109117773B (en) | Image feature point detection method, terminal device and storage medium | |
| US20200125876A1 (en) | Method and Device for License Plate Positioning | |
| WO2017045443A1 (en) | Image retrieval method and system | |
| CN106250838A (en) | vehicle identification method and system | |
| CN111597884A (en) | Facial action unit identification method and device, electronic equipment and storage medium | |
| WO2015195301A1 (en) | Obtaining structural information from images | |
| CN105144239A (en) | Image processing device, program, and image processing method | |
| CN104239879B (en) | The method and device of separating character | |
| US10860877B2 (en) | Logistics parcel picture processing method, device and system | |
| US20170323149A1 (en) | Rotation invariant object detection | |
| CN112085701A (en) | Face ambiguity detection method and device, terminal equipment and storage medium | |
| CN110163095B (en) | Loop detection method, loop detection device and terminal equipment | |
| CN104268864A (en) | Card edge extracting method and device | |
| CN111488810A (en) | Face recognition method and device, terminal equipment and computer readable medium | |
| CN108520263B (en) | Panoramic image identification method and system and computer storage medium | |
| CN113269752A (en) | Image detection method, device terminal equipment and storage medium | |
| KR20190011722A (en) | Estimation of human orientation in images using depth information |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180608 |
|
| RJ01 | Rejection of invention patent application after publication |