CN108197624A - The recognition methods of certificate image rectification and device, computer storage media - Google Patents
The recognition methods of certificate image rectification and device, computer storage media Download PDFInfo
- Publication number
- CN108197624A CN108197624A CN201810104835.XA CN201810104835A CN108197624A CN 108197624 A CN108197624 A CN 108197624A CN 201810104835 A CN201810104835 A CN 201810104835A CN 108197624 A CN108197624 A CN 108197624A
- Authority
- CN
- China
- Prior art keywords
- certificate image
- image
- certificate
- processing
- obtains
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- 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
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- 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
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/247—Aligning, centring, orientation detection or correction of the image by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Character Input (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of certificate image rectification recognition methods and device, computer storage media, wherein, method includes:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image;Slant correction is carried out to the smooth certificate image using Radon algorithms or Hough transform and obtains upright certificate image;Image enhancement processing is carried out to the upright certificate image using two-dimension fourier transform and obtains enhancing certificate image.The present invention accurately identifies inclination for OCR technique, distortion certificate image provides reliable guarantee, can effectively reduce the image caused by image capturing angle and papery certificate placement position and identify mistake.
Description
Technical field
The present invention relates to technical field of image processing more particularly to word correction recognition methods, are exactly specifically a kind of
The recognition methods of certificate image rectification and device, computer storage media.
Background technology
With the development of computer and network technology, exist largely to the paper media number such as ancient books, document, bill, business card
The demand of word, digitlization here is not limited solely to carry out electronic image using scanner or camera, more important
Be that these paper media files are converted into readable, editable electronic document to store, in order to realize this process, need
Pictograph identification is carried out to the electronic image that scans or acquire, in this case optical character identification (OCR) technology meet the tendency of and
It is raw.
OCR technique refers to that electronic equipment (for example, scanner or digital camera) checks the character printed on paper, then uses word
Shape is translated into the process of computword by symbol recognition methods;That is, be scanned to text information, then to image file into
Row analyzing and processing obtains the process of word and layout information.OCR technique is able to extensive use in production and living, Chinese
It is the features such as discrimination of block letter can reach more than 98%, the friendly of OCR technique, stability, ease for use and feasibility, deep
By Text region, worker is liked.
Current OCR technique can accurately handle the high-definition image (figure obtained from scanner obtained from scanner
Picture obtains, therefore hardly there are geometric distortions under conditions of source document is substantially flat), for being deposited in angle
In certain inclined image, existing OCR technique can barely cope with, but can't know for there are the images of geometric distortion
Not, main reason is that OCR technique effectively cannot rationally divide the image of geometric distortion, simple two can not be taken
Value method carries out feature extraction.In today that mobile terminal is prevailing, it is direct that people get used to the camera that mobile terminal is utilized to carry
Paper media is shot, such as people shoot the certificates such as identity card, business license using mobile terminal, are removed above these certificate images
Have except text information, also there is the non-legible information such as a large amount of mark, pictures, if using OCR technique Direct Recognition this
A little certificate images for tilting, distorting, it may appear that the phenomenon that identifying mistake or even None- identified.
Therefore, those skilled in the art are there is an urgent need for researching and developing a kind of certificate image rectification recognition methods, so as to the card that tilts, distorts
Book image precisely identifies.
Invention content
In view of this, the technical problem to be solved in the present invention is to provide a kind of certificate image rectification recognition methods and dress
It puts, computer storage media, solves the problems, such as that existing OCR technique can not correctly identify inclination, distortion certificate image.
In order to solve the above-mentioned technical problem, specific embodiment of the invention provides a kind of certificate image rectification identification side
Method, including:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image;Utilize Radon
Algorithm or Hough transform carry out slant correction to the smooth certificate image and obtain upright certificate image;Utilize two-dimension fourier
Transformation carries out image enhancement processing to the upright certificate image and obtains enhancing certificate image.
The specific embodiment of the present invention also provides a kind of certificate image rectification identification device, including:Fitting correction unit,
Smooth certificate image is obtained for being fitted correction to distortion certificate image using high order fitting function;Tilt corrector unit,
Upright certificate image is obtained for carrying out slant correction to the smooth certificate image using Radon algorithms or Hough transform;Figure
Image intensifying unit obtains enhancing card for carrying out image enhancement processing to the upright certificate image using two-dimension fourier transform
Book image.
The specific embodiment of the present invention also provides a kind of computer storage media for including computer executed instructions, described
When computer executed instructions are handled via data processing equipment, which performs certificate image rectification recognition methods.
Above-mentioned specific embodiment according to the present invention is it is found that the recognition methods of certificate image rectification and device, computer are deposited
Storage media at least has the advantages that:Correction is fitted certificate image using higher order functionality first, recycles Radon
Algorithm or Hough transform carry out slant correction to the certificate image after fitting correction, then using two-dimension fourier transform to inclining
Tiltedly the certificate image after correction carries out image enhancement processing, finally carries out gray scale to the certificate image after image enhancement processing
Change is handled.The invention enables digital camera, mobile terminals etc. to become effective information collection tool, is accurately identified for OCR technique
Inclination, distortion certificate image provide reliable guarantee, can effectively reduce since image capturing angle and papery certificate put position
Image identification mistake caused by putting.
It is to be understood that above-mentioned general description and detailed description below are merely illustrative and illustrative, not
It can the limitation range of the invention to be advocated.
Description of the drawings
Following appended attached drawing is the part of specification of the present invention, depicts example embodiments of the present invention, institute
Attached drawing is used for illustrating the principle of the present invention together with the description of specification.
Fig. 1 is the flow of the embodiment one of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure.
Fig. 2 is the flow of the embodiment two of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure.
Fig. 3 is the flow of the embodiment three of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure.
Fig. 4 is the flow of the example IV of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure.
Fig. 5 is the structure of the embodiment one of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram.
Fig. 6 is the structure of the embodiment two of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram.
Fig. 7 is the structure of the embodiment three of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram.
Fig. 8 is the structure of the example IV of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram.
Specific embodiment
Purpose, technical scheme and advantage to make the embodiment of the present invention are more clearly understood, below will with attached drawing and in detail
Narration clearly illustrates the spirit of disclosed content, and any skilled artisan is understanding the content of present invention
After embodiment, when the technology that can be taught by the content of present invention, it is changed and modifies, without departing from the essence of the content of present invention
God and range.
The illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but not as a limitation of the invention.
In addition, element/component of the same or like label used in drawings and the embodiments is for representing same or like portion
Point.
About " first " used herein, " second " ... etc., not especially censure the meaning of order or cis-position,
It is non-to limit the present invention, only for distinguishing the element described with same technique term or operation.
About direction term used herein, such as:Upper and lower, left and right, front or rear etc. are only the sides of refer to the attached drawing
To.Therefore, the direction term used is intended to be illustrative and not intended to limit this creation.
It is open term, i.e., about "comprising" used herein, " comprising ", " having ", " containing " etc.
Mean including but not limited to.
About it is used herein " and/or ", including any of the things or all combination.
Include " two " and " two or more " about " multiple " herein;Include " two groups " about " multigroup " herein
And " more than two ".
About term used herein " substantially ", " about " etc., to modify it is any can be with the quantity or mistake of microvariations
Difference, but this slight variations or error can't change its essence.In general, microvariations that such term is modified or error
Range in some embodiments can be 20%, in some embodiments can be 10%, can be in some embodiments 5% or its
His numerical value.It will be understood by those skilled in the art that the aforementioned numerical value referred to can be adjusted according to actual demand, it is not limited thereto.
It is certain to describe the word of the application by lower or discuss in the other places of this specification, to provide art technology
Personnel's guiding additional in relation to the description of the present application.
Fig. 1 is the flow of the embodiment one of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure as shown in Figure 1, being fitted correction, slant correction, image enhancement processing to distortion certificate image successively, obtains enhancing card
Book image.
In the specific embodiment shown in the drawings, certificate image rectification recognition methods includes:
Step 101:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image.This
In the specific embodiment of invention, distortion certificate image is referred specifically in certificate shooting process, and certain reasons lead to certificate scalloping
Distortion;Smooth certificate image refers specifically to the certificate image after correction distortion.
In specific embodiments of the present invention, the high order fitting function F is specially:
Wherein, K1Twist correcting parameter in horizontal direction in order to control;K2Twist correcting ginseng on vertical direction in order to control
Number;(u, v) is the coordinate system after certificate pattern distortion;(x, y) is the coordinate system before certificate pattern distortion.
Step 102:Slant correction acquisition side is carried out to the smooth certificate image using Radon algorithms or Hough transform
Positive certificate image.In specific embodiments of the present invention, smooth certificate image has certain angle of inclination, is unfavorable for the knowledge of later stage word
Not;Upright certificate image is the smooth certificate image eliminated after tilting.
Step 103:Carrying out image enhancement processing to the upright certificate image using two-dimension fourier transform is enhanced
Certificate image.In specific embodiments of the present invention, due to containing a large amount of useless and interference informations in upright certificate image, to Founder
Certificate image carries out image enhancement processing, can inhibit or remove these useless and interference informations, improve the visual effect of image.
Referring to Fig. 1, the processing such as correction, slant correction, image enhancement are fitted to distortion certificate image, are obtained clearly
Certificate image, for OCR technique accurately identify inclination, distortion certificate image provide reliable guarantee, can effectively reduce due to figure
The image as caused by acquisition angles and papery certificate placement position identifies mistake.
Fig. 2 is the flow of the embodiment two of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure as shown in Fig. 2, carrying out gray processing processing to enhancing certificate image using gray value divided linear strength algorithm, further carries
The accuracy of high OCR certificates identification.
In the specific embodiment shown in the drawings, certificate image rectification recognition methods further includes:
Step 104:Gray processing processing is carried out using gray value divided linear strength algorithm to the enhancing certificate image to obtain
Obtain gray scale certificate image.In specific embodiments of the present invention, gray processing processing is carried out using gray scale divided linear strength function.
Referring to Fig. 2, the purpose of gray processing processing is carried out to enhancing certificate image using gray value divided linear strength algorithm
So that the gray scale between the high background information of the low font information of gray value, the noise signal of gray value random distribution, gray value
Distance is stretched, and in order to selected threshold during binary conversion treatment, is carried out gray processing processing to enhancing certificate image, is further carried
The accuracy of high OCR certificates identification.
Fig. 3 is the flow of the embodiment three of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure, as shown in figure 3, carrying out gray processing processing to enhancing certificate image using gray value divided linear strength algorithm obtains gray scale card
Book image.
In the specific embodiment shown in the drawings, step 104 specifically includes:
Step 1041:The gray value of the enhancing certificate image is divided into multiple gray scale intervals.The specific implementation of the present invention
In example, the gray value for enhancing certificate image is divided into 3 gray scale intervals.
Step 1042:According to the color distribution feature of the enhancing certificate image respectively to the corresponding gray scale interval point
With weights.In specific embodiments of the present invention, color distribution feature includes pigment and brightness etc.;Pigment more multiple weighing value is bigger, bright
Degree is bigger, and weights are also bigger, for example, for RGB image, red pigments are more, and weights are bigger;Brightness of image is bigger,
Weights are also bigger.
Step 1043:The gray scale interval is extended using the weights and obtains the gray scale certificate image.This hair
In bright specific embodiment, corresponding gray scale interval is multiplied by using weights, obtains new gray scale interval.
Referring to Fig. 3, gray processing processing is carried out to enhancing certificate image using gray value divided linear strength algorithm, convenient for two
Selected threshold when value is handled carries out gray processing processing to enhancing certificate image, further improves the accurate of OCR certificates identification
Property.
Fig. 4 is the flow of the example IV of a kind of certificate image rectification recognition methods that the specific embodiment of the invention provides
Figure as shown in figure 4, carrying out image enhancement processing to upright certificate graphs picture using two-dimension fourier transform, can inhibit or remove
Useless and interference information in upright certificate image improves the visual effect of image.
In the specific embodiment shown in the drawings, step 103 specifically includes:
Step 1031:Two-dimension fourier transform is carried out to the upright certificate image and obtains two-dimensional frequency image.
Step 1032:Corresponding frequency domain filter is determined according to the two-dimensional frequency image.Specific embodiments of the present invention
In, the high frequency waves distribution situation in two-dimensional frequency image determines corresponding frequency domain filter.
Step 1033:The upright certificate image, which is handled, using the frequency domain filter obtains the enhancing certificate image.
In specific embodiments of the present invention, the high frequency waves signal in upright certificate image is filtered out using frequency domain filter, after raising
The accuracy of phase OCR certificate identification.
Referring to Fig. 4, image enhancement processing is carried out to upright certificate graphs picture using two-dimension fourier transform, can inhibit or go
Except the useless and interference information in upright certificate image, improve the visual effect of image, improve the accurate of later stage OCR certificate identification
Property.
Fig. 5 is the structure of the embodiment one of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram, device as shown in Figure 5 can be applied in Fig. 1~method shown in Fig. 4, and distortion certificate image is intended successively
Correction, slant correction, image enhancement processing are closed, obtains enhancing certificate image.
In the specific embodiment shown in the drawings, certificate image rectification identification device includes:Fitting correction unit 1 inclines
Tiltedly correction unit 2 and image enhancing unit 3.Wherein, fitting correction unit 1 is used for using high order fitting function to the certificate graphs that distort
Smooth certificate image is obtained as being fitted correction;Tilt corrector unit 2 is used for using Radon algorithms or Hough transform to institute
It states smooth certificate image and carries out the upright certificate image of slant correction acquisition;Image enhancing unit 3 is used to become using two-dimension fourier
It changes and image enhancement processing acquisition enhancing certificate image is carried out to the upright certificate image.
Referring to Fig. 5, the processing such as correction, slant correction, image enhancement are fitted to distortion certificate image, are obtained clearly
Certificate image, for OCR technique accurately identify inclination, distortion certificate image provide reliable guarantee, can effectively reduce due to figure
The image as caused by acquisition angles and papery certificate placement position identifies mistake.
Fig. 6 is the structure of the embodiment two of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram, as shown in fig. 6, gray processing processing unit carries out ash using gray value divided linear strength algorithm to enhancing certificate image
Degreeization processing further improves the accuracy of OCR certificates identification.
In the specific embodiment shown in the drawings, certificate image rectification identification device further includes gray processing processing unit 4.
Wherein, gray processing processing unit 4 is used to carry out gray scale to the enhancing certificate image using gray value divided linear strength algorithm
Change processing and obtain gray scale certificate image.
Referring to Fig. 6, the purpose of gray processing processing is carried out to enhancing certificate image using gray value divided linear strength algorithm
So that the gray scale between the high background information of the low font information of gray value, the noise signal of gray value random distribution, gray value
Distance is stretched, and in order to selected threshold during binary conversion treatment, is carried out gray processing processing to enhancing certificate image, is further carried
The accuracy of high OCR certificates identification.
Fig. 7 is the structure of the embodiment three of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram, as shown in fig. 7, gray processing processing unit specifically includes division module, distribution module and expansion module.
In the specific embodiment shown in the drawings, the gray processing processing unit 4 specifically includes division module 41, distribution
Module 42 and expansion module 43.Wherein, division module 41 is used to the gray value of the enhancing certificate image being divided into multiple gray scales
Section;Distribution module 42 is used for the color distribution feature according to the enhancing certificate image respectively to the corresponding gray scale interval
Distribute weights;Expansion module 43, which is used to be extended the gray scale interval using the weights, obtains the gray scale certificate graphs
Picture.
Referring to Fig. 7, gray processing processing is carried out to enhancing certificate image using gray value divided linear strength algorithm, convenient for two
Selected threshold when value is handled carries out gray processing processing to enhancing certificate image, further improves the accurate of OCR certificates identification
Property.
Fig. 8 is the structure of the example IV of a kind of certificate image rectification identification device that the specific embodiment of the invention provides
Schematic diagram, as shown in figure 8, using two-dimension fourier transform to upright certificate graphs picture carry out image enhancement processing, can inhibit or
Useless and interference information in the upright certificate image of removal improves the visual effect of image.
In the specific embodiment shown in the drawings, described image enhancement unit 3 specifically includes:Conversion module 31 determines
Module 32 and processing module 33.Wherein, conversion module 31 is used to obtain the upright certificate image progress two-dimension fourier transform
To two-dimensional frequency image;Determining module 32 is used to determine corresponding frequency domain filter according to the two-dimensional frequency image;Handle mould
Block 33 is used to obtain the enhancing certificate image using the frequency domain filter processing upright certificate image.
Referring to Fig. 8, image enhancement processing is carried out to upright certificate graphs picture using two-dimension fourier transform, can inhibit or go
Except the useless and interference information in upright certificate image, improve the visual effect of image, improve the accurate of later stage OCR certificate identification
Property.
The specific embodiment of the invention provides a kind of computer storage media for including computer executed instructions, the calculating
When machine execute instruction is handled via data processing equipment, which performs certificate image rectification recognition methods.Method
It may comprise steps of:
Step 101:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image.
Step 102:Slant correction acquisition side is carried out to the smooth certificate image using Radon algorithms or Hough transform
Positive certificate image.
Step 103:Carrying out image enhancement processing to the upright certificate image using two-dimension fourier transform is enhanced
Certificate image.
The specific embodiment of the invention also provides a kind of computer storage media for including computer executed instructions, the meter
When calculation machine execute instruction is handled via data processing equipment, which performs certificate image rectification recognition methods.Side
Method may comprise steps of:
Step 101:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image.
Step 102:Slant correction acquisition side is carried out to the smooth certificate image using Radon algorithms or Hough transform
Positive certificate image.
Step 103:Carrying out image enhancement processing to the upright certificate image using two-dimension fourier transform is enhanced
Certificate image.
Step 104:Gray processing processing is carried out using gray value divided linear strength algorithm to the enhancing certificate image to obtain
Obtain gray scale certificate image.
The specific embodiment of the invention also provides a kind of computer storage media for including computer executed instructions, the meter
When calculation machine execute instruction is handled via data processing equipment, which performs certificate image rectification recognition methods.Side
Method may comprise steps of:
Step 101:Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image.
Step 102:Slant correction acquisition side is carried out to the smooth certificate image using Radon algorithms or Hough transform
Positive certificate image.
Step 103:Carrying out image enhancement processing to the upright certificate image using two-dimension fourier transform is enhanced
Certificate image.
Step 1041:The gray value of the enhancing certificate image is divided into multiple gray scale intervals.
Step 1042:According to the color distribution feature of the enhancing certificate image respectively to the corresponding gray scale interval point
With weights.
Step 1043:The gray scale interval is extended using the weights and obtains the gray scale certificate image.
The specific embodiment of the invention also provides a kind of computer storage media for including computer executed instructions, the meter
When calculation machine execute instruction is handled via data processing equipment, which performs certificate image rectification recognition methods.Side
Method may comprise steps of:
Step 1031:Two-dimension fourier transform is carried out to the upright certificate image and obtains two-dimensional frequency image.
Step 1032:Corresponding frequency domain filter is determined according to the two-dimensional frequency image.
Step 1033:The upright certificate image, which is handled, using the frequency domain filter obtains the enhancing certificate image.
The specific embodiment of the invention provides a kind of certificate image rectification recognition methods and device, computer storage media, first
Correction is first fitted certificate image using higher order functionality, after recycling Radon algorithms or Hough transform to fitting correction
Certificate image carries out slant correction, then carries out image enhancement to the certificate image after slant correction using two-dimension fourier transform
Processing finally carries out gray processing processing to the certificate image after image enhancement processing.The invention enables digital camera, movements
Terminal etc. becomes effective information collection tool, accurately identifies inclination for OCR technique, distortion certificate image provides reliable guarantor
Barrier can effectively reduce the image caused by image capturing angle and papery certificate placement position and identify mistake.
The above-mentioned embodiment of the present invention can be implemented in various hardware, Software Coding or both combination.For example, this hair
Bright embodiment, which is alternatively in data signal processor (Digital Signal Processor, DSP), performs the above method
Program code.The present invention can also refer to computer processor, digital signal processor, microprocessor or field-programmable gate array
Arrange the multiple functions that (Field Programmable Gate Array, FPGA) is performed.Can above-mentioned processing be configured according to the present invention
Device performs particular task, and the machine-readable software code of ad hoc approach or the firmware generation that the present invention discloses are defined by performing
Code is completed.Software code or firmware code can be developed into different program languages and different forms or form.Or
Different target platform composing software codes.However, in generation, is configured according to the software code of execution task of the present invention and other types
Different code pattern, type and the language of code do not depart from spirit and scope of the invention.
The foregoing is merely the schematical specific embodiment of the present invention, before the design of the present invention and principle is not departed from
It puts, the equivalent variations and modification that any those skilled in the art is made should all belong to the scope of protection of the invention.
Claims (10)
1. a kind of certificate image rectification recognition methods, which is characterized in that this method includes:
Correction is fitted to distortion certificate image using high order fitting function and obtains smooth certificate image;
Slant correction is carried out to the smooth certificate image using Radon algorithms or Hough transform and obtains upright certificate image;With
And
Image enhancement processing is carried out to the upright certificate image using two-dimension fourier transform and obtains enhancing certificate image.
2. certificate image rectification recognition methods as described in claim 1, which is characterized in that this method further includes:
Gray processing processing is carried out to the enhancing certificate image using gray value divided linear strength algorithm and obtains gray scale certificate graphs
Picture.
3. certificate image rectification recognition methods as claimed in claim 2, which is characterized in that using gray value divided linear strength
Algorithm carries out the step of gray processing processing obtains gray scale certificate image to the enhancing certificate image, specifically includes:
The gray value of the enhancing certificate image is divided into multiple gray scale intervals;
According to the color distribution feature of the enhancing certificate image respectively to the corresponding gray scale interval distribution weights;And
The gray scale interval is extended using the weights and obtains the gray scale certificate image.
4. certificate image rectification recognition methods as described in claim 1, which is characterized in that using two-dimension fourier transform to institute
It states upright certificate image and carries out the step of image enhancement processing obtains enhancing certificate image, specifically include:
Two-dimension fourier transform is carried out to the upright certificate image and obtains two-dimensional frequency image;
Corresponding frequency domain filter is determined according to the two-dimensional frequency image;And
The upright certificate image, which is handled, using the frequency domain filter obtains the enhancing certificate image.
5. certificate image rectification recognition methods as described in claim 1, which is characterized in that the high order fitting function F is specific
For:
Wherein, K1Twist correcting parameter in horizontal direction in order to control;K2Twist correcting parameter on vertical direction in order to control;(u,
V) it is the coordinate system after certificate pattern distortion;(x, y) is the coordinate system before certificate pattern distortion.
6. a kind of certificate image rectification identification device, which is characterized in that the device includes:
Fitting correction unit obtains smooth certificate graphs for being fitted correction to distortion certificate image using high order fitting function
Picture;
Tilt corrector unit obtains for carrying out slant correction to the smooth certificate image using Radon algorithms or Hough transform
Obtain upright certificate image;And
Image enhancing unit, for carrying out image enhancement processing acquisition to the upright certificate image using two-dimension fourier transform
Enhance certificate image.
7. certificate image rectification identification device as claimed in claim 6, which is characterized in that the device further includes:
Gray processing processing unit, for carrying out gray processing to the enhancing certificate image using gray value divided linear strength algorithm
Processing obtains gray scale certificate image.
8. certificate image rectification identification device as claimed in claim 7, which is characterized in that the gray processing processing unit is specific
Including:
Division module, for the gray value of the enhancing certificate image to be divided into multiple gray scale intervals;
Distribution module, for being divided respectively to the corresponding gray scale interval according to the color distribution feature of the enhancing certificate image
With weights;And
Expansion module obtains the gray scale certificate image for being extended using the weights to the gray scale interval.
9. certificate image rectification identification device as claimed in claim 7, which is characterized in that described image enhancement unit is specifically wrapped
It includes:
Conversion module obtains two-dimensional frequency image for carrying out two-dimension fourier transform to the upright certificate image;
Determining module, for determining corresponding frequency domain filter according to the two-dimensional frequency image;And
Processing module, for obtaining the enhancing certificate image using the frequency domain filter processing upright certificate image.
10. a kind of computer storage media for including computer executed instructions, the computer executed instructions are via data processing
During equipment processing, which performs the certificate image rectification recognition methods as described in Claims 1 to 5 is any.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810104835.XA CN108197624A (en) | 2018-02-02 | 2018-02-02 | The recognition methods of certificate image rectification and device, computer storage media |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810104835.XA CN108197624A (en) | 2018-02-02 | 2018-02-02 | The recognition methods of certificate image rectification and device, computer storage media |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108197624A true CN108197624A (en) | 2018-06-22 |
Family
ID=62591952
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810104835.XA Pending CN108197624A (en) | 2018-02-02 | 2018-02-02 | The recognition methods of certificate image rectification and device, computer storage media |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108197624A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109447067A (en) * | 2018-10-24 | 2019-03-08 | 北方民族大学 | A kind of bill angle detecting antidote and automatic ticket checking system |
CN109614972A (en) * | 2018-12-06 | 2019-04-12 | 泰康保险集团股份有限公司 | Image processing method, device, electronic equipment and computer-readable medium |
CN112733831A (en) * | 2021-01-08 | 2021-04-30 | 南昌航空大学 | Trademark detection method and system |
CN113554558A (en) * | 2020-04-26 | 2021-10-26 | 北京金山数字娱乐科技有限公司 | Image processing method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411775A (en) * | 2011-08-10 | 2012-04-11 | Tcl集团股份有限公司 | Method and system for correcting brightness of text image |
US20130201358A1 (en) * | 2012-02-06 | 2013-08-08 | Apple Inc. | Efficient Line Detection Method |
CN104867113A (en) * | 2015-03-31 | 2015-08-26 | 酷派软件技术(深圳)有限公司 | Method and system for perspective distortion correction of image |
CN106778752A (en) * | 2016-11-16 | 2017-05-31 | 广西大学 | A kind of character recognition method |
CN106951896A (en) * | 2017-02-22 | 2017-07-14 | 武汉黄丫智能科技发展有限公司 | A kind of license plate image sloped correcting method |
-
2018
- 2018-02-02 CN CN201810104835.XA patent/CN108197624A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102411775A (en) * | 2011-08-10 | 2012-04-11 | Tcl集团股份有限公司 | Method and system for correcting brightness of text image |
US20130201358A1 (en) * | 2012-02-06 | 2013-08-08 | Apple Inc. | Efficient Line Detection Method |
CN104867113A (en) * | 2015-03-31 | 2015-08-26 | 酷派软件技术(深圳)有限公司 | Method and system for perspective distortion correction of image |
CN106778752A (en) * | 2016-11-16 | 2017-05-31 | 广西大学 | A kind of character recognition method |
CN106951896A (en) * | 2017-02-22 | 2017-07-14 | 武汉黄丫智能科技发展有限公司 | A kind of license plate image sloped correcting method |
Non-Patent Citations (2)
Title |
---|
ZHENGJIANGTAO0: "OCR系统中的图像预处理的研究和实现", 《百度文库》 * |
陈爱军,董光辉,夏宇: "《数字图像处理及其MATLAB实现》", 31 July 2008 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109447067A (en) * | 2018-10-24 | 2019-03-08 | 北方民族大学 | A kind of bill angle detecting antidote and automatic ticket checking system |
CN109447067B (en) * | 2018-10-24 | 2021-07-20 | 北方民族大学 | Bill direction detection and correction method and automatic bill checking system |
CN109614972A (en) * | 2018-12-06 | 2019-04-12 | 泰康保险集团股份有限公司 | Image processing method, device, electronic equipment and computer-readable medium |
CN113554558A (en) * | 2020-04-26 | 2021-10-26 | 北京金山数字娱乐科技有限公司 | Image processing method and device |
CN112733831A (en) * | 2021-01-08 | 2021-04-30 | 南昌航空大学 | Trademark detection method and system |
CN112733831B (en) * | 2021-01-08 | 2022-04-15 | 南昌航空大学 | Trademark detection method and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9230382B2 (en) | Document image capturing and processing | |
RU2678485C1 (en) | Method of character segmentation and recognition | |
CN104112128B (en) | Digital image processing system and method applied to bill image character recognition | |
CN108197624A (en) | The recognition methods of certificate image rectification and device, computer storage media | |
US8577099B2 (en) | Method, apparatus, and program for detecting facial characteristic points | |
EP2973226A1 (en) | Classifying objects in digital images captured using mobile devices | |
WO2007127085A1 (en) | Generating a bitonal image from a scanned colour image | |
CN111353961B (en) | Document curved surface correction method and device | |
CN109558844A (en) | The method and apparatus of self-defined template discrimination is promoted based on image normalization | |
CN108985305B (en) | Laser etching industrial detonator coded image positioning and correcting method | |
CN108537787B (en) | Quality judgment method for face image | |
CN109784342A (en) | A kind of OCR recognition methods and terminal based on deep learning model | |
CN110598566A (en) | Image processing method, device, terminal and computer readable storage medium | |
CN111222507B (en) | Automatic identification method for digital meter reading and computer readable storage medium | |
CN110738216A (en) | Medicine identification method based on improved SURF algorithm | |
CN113903024A (en) | Handwritten bill numerical value information identification method, system, medium and device | |
CN105225218A (en) | For distortion correction method and the equipment of file and picture | |
CN113158895A (en) | Bill identification method and device, electronic equipment and storage medium | |
CN109741273A (en) | A kind of mobile phone photograph low-quality images automatically process and methods of marking | |
CN116052152A (en) | License plate recognition system based on contour detection and deep neural network | |
CN111582115B (en) | Financial bill processing method, device, equipment and readable storage medium | |
JP2018137636A (en) | Image processing device and image processing program | |
Pan et al. | An efficient method for skew correction of license plate | |
CN115410191B (en) | Text image recognition method, device, equipment and storage medium | |
WO2015092059A1 (en) | Method and system for correcting projective distortions. |
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: 20180622 |
|
RJ01 | Rejection of invention patent application after publication |