CN109359497A - A kind of positioning and recognition methods in VAT invoice two-dimensional image code region - Google Patents
A kind of positioning and recognition methods in VAT invoice two-dimensional image code region Download PDFInfo
- Publication number
- CN109359497A CN109359497A CN201811081051.6A CN201811081051A CN109359497A CN 109359497 A CN109359497 A CN 109359497A CN 201811081051 A CN201811081051 A CN 201811081051A CN 109359497 A CN109359497 A CN 109359497A
- Authority
- CN
- China
- Prior art keywords
- image
- identification
- region
- candidate
- recognition result
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/146—Methods for optical code recognition the method including quality enhancement steps
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Abstract
The invention discloses the positioning and recognition methods in a kind of VAT invoice two-dimensional image code region, including carry out locating segmentation to VAT invoice two-dimensional image code region and identify to two dimensional code；The present invention carries out locating segmentation to the two-dimension code area in invoice picture when carrying out OCR identification and does enhancing processing, finally it is sent into identification, improve the performance of two dimensional code identification, invoice information is inquired according to two dimensional code recognition result to supplement the OCR result identified, improve OCR accuracy of identification, artificial investment is reduced, is reduced costs for enterprise.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of positioning in VAT invoice two-dimensional image code region
With recognition methods.
Background technique
In bill OCR identification especially VAT invoice identification, the two-dimensional barcode information in invoice contains the one of invoice
A little critical fielies, such as invoice number, invoice codes, the amount of money, make out an invoice the date.And financial system automate book keeping operation in terms of need by
Paper invoice is scanned into electronic pictures, uses OCR technique to identify invoice image content then to reach automation book keeping operation
Purpose.However, causing the OCR accuracy of identification of bill limited, due to various reasons in order to improve the OCR identification essence of invoice picture
Degree also can yet be regarded as by the two-dimensional barcode information in bill and a kind of effectively may be used to reduce the artificial investment checked recognition result
Capable means.In the identification application of existing two dimensional code, generally it is both needed to camera and is identified and done acquiring image in 2 D code
Subsequent business operation, and the application of the two dimensional code in the existing picture of Direct Recognition (mobile phone photograph, scanner scanning) is few, and
There is a higher requirement to picture quality, for example picture illumination light and shade unevenness, torsional deformation, noise are serious etc. to two-dimension code area
Positioning influences serious with identification.
Summary of the invention
To solve problems of the prior art, the object of the present invention is to provide a kind of VAT invoice two-dimensional image codes
The positioning and recognition methods in region；The present invention carries out positioning point to the two-dimension code area in invoice picture when carrying out OCR identification
It cuts and does enhancing processing, be finally sent into identification, improve the performance of two dimensional code identification, invoice is inquired according to two dimensional code recognition result
Information supplements the OCR result identified, improves OCR accuracy of identification, reduces artificial investment, reduces costs for enterprise.
To achieve the above object, the technical solution adopted by the present invention is that: a kind of VAT invoice two-dimensional image code region
Positioning and recognition methods, including locating segmentation is carried out to VAT invoice two-dimensional image code region and two dimensional code is known
Not；Wherein, to VAT invoice two-dimensional image code region carry out locating segmentation the following steps are included:
A, the image uploaded to client pre-processes, and the pretreatment includes doing shadow removal to image, removing black surround
And enhancing picture contrast；
B, the continuous black pixel point region at removal image four edges up and down；
C, the noise spot in Gaussian Blur processing removal image is done to the image after step B processing；
D, OTSU binary conversion treatment is done to the image after step C processing, and judges after OTSU binary conversion treatment
Image it is whether qualified: as qualified, be then directly entered step E, it is such as unqualified, then using self-adaption binaryzation to image at
Step E is entered back into after reason；When the surface of VAT invoice picture has uneven illumination, self-adaption binaryzation handles phase
The shade in image can be more effectively removed for OTSU binary conversion treatment, therefore the present invention uses OTSU binary conversion treatment and OTSU
The method that both binary conversion treatments combine；
E, will be by step D treated image scaling to uniform sizes, choosing sizeable all black picture region is
Template carries out traversing to match doing convolution algorithm, obtains calculated matrix of consequence with the image after scaling；
F, it according to the threshold value mixed up, is taken out from matrix of consequence using non-maxima suppression algorithm with template matching effect most
Good candidate region；
G, it is associated with the candidate region of intersection；
H, non-maxima suppression algorithm is executed to the candidate region after merging again and obtains final candidate two-dimension code area；
Two dimensional code is identified the following steps are included:
A, original image is sent into zxing identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step b；
B, original image is sent into zbarlight identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step
c；
C, it will be sent into zxing identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise continue to hold
Row step d；
D, it will be sent into zbarlight identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise after
It is continuous to execute step e；
E, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zxing identification, if identifying successfully, returned
Return recognition result；Otherwise continue to execute step f；
F, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zbarlight identification, if being identified as
Function returns to recognition result；Otherwise continue to execute step g；
G, zxing identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H, if knowing
Not Cheng Gong, return recognition result；Otherwise continue to execute step h；
H, zbarlight identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H,
If identifying successfully, recognition result is returned；Otherwise recognition failures are returned, identification process is terminated.
Further, in the step D, judge that whether qualified the standard of the image after OTSU binary conversion treatment be as follows:Wherein: NOB indicates the number of the black pixel point in image after OTSU binaryzation, and W is
The width of OTSU binary image, H are the height of OTSU binary image, and threshold is the threshold value of setting.
Further, in the step E, if the width of selected template and high respectively W, h, specifically traverse matching process
It is as follows: R (i, j)=∑I, j(T (m, n)-S (i+m, j+n))2, in which: m=0,1,2 ..., w-1；N=0,1,2 ..., h-1；T
(m, n) is the pixel in template, and S (i+m, j+n) is the pixel in original image, obtains (W-w+1, H-h+1) according to above formula
Matrix of consequence.
The following further describes the technical solution of the present invention below:
Prior art arrangement there are aiming at the problem that, it may be assumed that without using camera acquire image in 2 D code and directly handle packet
Picture containing two dimensional code is more demanding to picture quality, it is difficult to be accurately positioned from picture and be partitioned into two-dimension code area；It is existing
Two dimensional code recognizer has respective advantage and disadvantage, if can do in the application to it to merge, in conjunction with respective advantage, then to mentioning
Rising two dimensional code discrimination has certain help.
For the QR code content being recognized accurately in VAT invoice picture, the present invention starts with processing in terms of three: one,
Two-dimension code area is accurately positioned out from picture；Two, image enhancement processing is done to the two-dimension code image being partitioned into from original image, with
The details for removing noise jamming as far as possible, restoring image；Three, a variety of recognition strategies are merged, the recognition performance of two dimensional code algorithm is promoted.
Specific technical solution is illustrated for above 3 points separately below:
1) the two-dimension code area feature for analyzing invoice image, it is found that the energy of two-dimension code area is more concentrated.Therefore, subject to
It really is partitioned into two-dimension code area from invoice picture, using the method for template matching.First extremely by VAT invoice image scaling
Uniform sizes, choosing sizeable all black picture region is that the binary image after template, with scaling does convolution algorithm, is selected
Take suitable threshold value to after convolution result filtered, finally using NMS (non-maxima suppression algorithm) merge candidate regions obtain
To final two-dimension code area.
2) picture quality is also to influence an important factor for two dimensional code positioning is with identification.In practical business, part picture is found
There is black surround and there may be the phenomenons of light and shade unevenness, in order to mitigate image border black surround and image light and shade unevenness to two-dimension code area
The influence of segmentation need to do the operation of removal black surround, and use self-adaption binaryzation method before template matching to image border
To remove the shade in image, so that image shade be avoided to cross and influence the positioning of two-dimension code area.In addition, in practical business
It is that scanner scanning is got, therefore is limited to the precision of scanner that invoice picture, which also has larger a part, the two dimension in picture
The problem of code region is lost there may be partial region pixel, to solve this problem, this method for that can not know under normal conditions
Other two dimensional code does the processing of Gaussian Blur, to restore the two-dimension code area being truncated to as far as possible.In addition, the image that actual acquisition arrives
May also be by noise jamming, therefore the processing for removing dry sound can also be done to the image in 2 D code being partitioned into, it is then re-fed into identification,
To improve discrimination.
3) in view of the recognition effect of different two dimensional code recognizer packets is different, using the multiple identifications of fusion in this method
The strategy of algorithm is respectively adopted zxing and zbarlight, first send the image in 2 D code after aforesaid operations processing
Enter zxing to be identified, be exited if identifying successfully.Otherwise, continue to be identified using zbarlight, and return finally
Recognition result.Difference of the two in recognizer can effectively be combined using the strategy, improve discrimination.
The beneficial effects of the present invention are: the present invention can be accurately positioned the two-dimension code area in VAT invoice picture, and
The two dimensional code being partitioned into is identified, in the case where invoice picture quality is met certain condition, (two-dimension code area is without serious
Torsional deformation, two-dimension code area are complete, unobstructed), the image in 2 D code being partitioned into can be accurately identified.
Detailed description of the invention
Fig. 1 is the flow diagram that the embodiment of the present invention carries out locating segmentation to VAT invoice two-dimensional image code region；
Fig. 2 is the flow diagram that the embodiment of the present invention identifies two dimensional code.
Specific embodiment
The embodiment of the present invention is described in detail with reference to the accompanying drawing.
Embodiment
A kind of positioning and recognition methods in VAT invoice two-dimensional image code region, including to VAT invoice two-dimensional image
Code region carries out locating segmentation and identifies to two dimensional code；Wherein, as shown in FIG. 1, FIG. 1 is the present embodiment sends out value-added tax
Ticket two-dimensional image code region carries out the flow diagram of locating segmentation, specifically includes the following steps:
A, the image uploaded to client pre-processes, and the pretreatment includes doing shadow removal to image, removing black surround
And enhancing picture contrast；
B, the continuous black pixel point region at removal image four edges up and down；
C, the noise spot in Gaussian Blur processing removal image is done to the image after step B processing；
D, OTSU binary conversion treatment is done to the image after step C processing, and judges after OTSU binary conversion treatment
Image it is whether qualified: as qualified, be then directly entered step E, it is such as unqualified, then using self-adaption binaryzation to image at
Step E is entered back into after reason；Judge that whether qualified the standard of the image after OTSU binary conversion treatment be as follows:Wherein: NOB indicates the number of the black pixel point in image after OTSU binaryzation, and W is
The width of OTSU binary image, H are the height of OTSU binary image, and threshold is the threshold value of setting；
E, will be by step D treated image scaling to uniform sizes, choosing sizeable all black picture region is
Template carries out traversing to match doing convolution algorithm, obtains calculated matrix of consequence with the image after scaling；If selected template
Wide and high respectively w, h, it is as follows specifically to traverse matching process: R (i, j)=∑I, j(T (m, n)-S (i+m, j+n))2, wherein:
M=0,1,2 ..., w-1；N=0,1,2 ..., h-1；T (m, n) is the pixel in template, and S (i+m, j+n) is the picture in original image
Vegetarian refreshments obtains the matrix of consequence of (W-w+1, H-h+1) according to above formula；
F, it according to the threshold value mixed up, is taken out from matrix of consequence using non-maxima suppression algorithm with template matching effect most
Good candidate region；
G, it is associated with the candidate region of intersection；
H, non-maxima suppression algorithm is executed to the candidate region after merging again and obtains final candidate two-dimension code area；
As shown in Fig. 2, Fig. 2 is the flow diagram that is identified to two dimensional code of the present embodiment, specifically includes the following steps:
A, original image is sent into zxing identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step b；
B, original image is sent into zbarlight identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step
c；
C, it will be sent into zxing identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise continue to hold
Row step d；
D, it will be sent into zbarlight identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise after
It is continuous to execute step e；
E, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zxing identification, if identifying successfully, returned
Return recognition result；Otherwise continue to execute step f；
F, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zbarlight identification, if being identified as
Function returns to recognition result；Otherwise continue to execute step g；
G, zxing identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H, if knowing
Not Cheng Gong, return recognition result；Otherwise continue to execute step h；
H, zbarlight identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H,
If identifying successfully, recognition result is returned；Otherwise recognition failures are returned, identification process is terminated.
A specific embodiment of the invention above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
Limitations on the scope of the patent of the present invention therefore cannot be interpreted as.It should be pointed out that for those of ordinary skill in the art
For, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to guarantor of the invention
Protect range.
Claims (3)
1. a kind of positioning and recognition methods in VAT invoice two-dimensional image code region, which is characterized in that including sending out value-added tax
Ticket two-dimensional image code region carries out locating segmentation and identifies to two dimensional code；Wherein, to VAT invoice two-dimensional image code
Region carry out locating segmentation the following steps are included:
A, to client upload image pre-process, it is described pretreatment include shadow removal is done to image, go black surround and
Enhance picture contrast；
B, the continuous black pixel point region at removal image four edges up and down；
C, the noise spot in Gaussian Blur processing removal image is done to the image after step B processing；
D, OTSU binary conversion treatment is done to the image after step C processing, and judges the figure after OTSU binary conversion treatment
Seem no qualification: such as qualification, being then directly entered step E, it is such as unqualified, then after being handled using self-adaption binaryzation image
Enter back into step E；
E, will be by step D treated image scaling to uniform sizes, choosing sizeable all black picture region is template,
It carries out traversing to match with the image after scaling doing convolution algorithm, obtains calculated matrix of consequence；
F, it according to the threshold value mixed up, is taken out from matrix of consequence using non-maxima suppression algorithm best with template matching effect
Candidate region；
G, it is associated with the candidate region of intersection；
H, non-maxima suppression algorithm is executed to the candidate region after merging again and obtains final candidate two-dimension code area；
Two dimensional code is identified the following steps are included:
A, original image is sent into zxing identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step b；
B, original image is sent into zbarlight identification, if identifying successfully, returns to recognition result；Otherwise continue to execute step c；
C, it will be sent into zxing identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise continue to execute step
Rapid d；
D, it will be sent into zbarlight identification after original image binaryzation, if identifying successfully, returns to recognition result；Otherwise continue to hold
Row step e；
E, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zxing identification, if identifying successfully, returns and know
Other result；Otherwise continue to execute step f；
F, the candidate two-dimension code area original image being partitioned into step H is sequentially sent to zbarlight identification, if identifying successfully, returned
Return recognition result；Otherwise continue to execute step g；
G, zxing identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H, if being identified as
Function returns to recognition result；Otherwise continue to execute step h；
H, zbarlight identification will be sequentially sent to after the candidate two-dimension code area original image binaryzation being partitioned into step H, if knowing
Not Cheng Gong, return recognition result；Otherwise recognition failures are returned, identification process is terminated.
2. the positioning and recognition methods in VAT invoice two-dimensional image code region according to claim 1, which is characterized in that
In the step D, judge that whether qualified the standard of the image after OTSU binary conversion treatment be as follows:Wherein: NOB indicates the number of the black pixel point in image after OTSU binaryzation, and W is
The width of OTSU binary image, H are the height of OTSU binary image, and threshold is the threshold value of setting.
3. the positioning and recognition methods in VAT invoice two-dimensional image code region according to claim 2, which is characterized in that
In the step E, if the wide and high of selected template is respectively w, h, specific traversal matching process is as follows: R (i, j)=∑I, j(T
(m, n)-S (i+m, j+n))2, in which: m=0,1,2 ..., w-1；N=0,1,2 ..., h-1；T (m, n) is the pixel in template
Point, S (i+m, j+n) are the pixel in original image, obtain the matrix of consequence of (W-w+1, H-h+1) according to above formula.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811081051.6A CN109359497A (en) | 2018-09-17 | 2018-09-17 | A kind of positioning and recognition methods in VAT invoice two-dimensional image code region |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811081051.6A CN109359497A (en) | 2018-09-17 | 2018-09-17 | A kind of positioning and recognition methods in VAT invoice two-dimensional image code region |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109359497A true CN109359497A (en) | 2019-02-19 |
Family
ID=65350887
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811081051.6A Pending CN109359497A (en) | 2018-09-17 | 2018-09-17 | A kind of positioning and recognition methods in VAT invoice two-dimensional image code region |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109359497A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222547A (en) * | 2019-04-29 | 2019-09-10 | 北京三快在线科技有限公司 | Bar code recognition and device |
| CN113643274A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Method and device for screening two-dimensional code candidate area |
| CN114092945A (en) * | 2022-01-21 | 2022-02-25 | 广州文石信息科技有限公司 | Handwriting input method and device based on ink screen |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254144A (en) * | 2011-07-12 | 2011-11-23 | 四川大学 | Robust method for extracting two-dimensional code area in image |
| CN102693409A (en) * | 2012-05-18 | 2012-09-26 | 四川大学 | Method for quickly identifying two-dimension code system type in images |
| CN103500323A (en) * | 2013-09-18 | 2014-01-08 | 西安理工大学 | Template matching method based on self-adaptive gray-scale image filtering |
| CN105976001A (en) * | 2016-05-19 | 2016-09-28 | 杭州博安实业有限公司 | False-proofing license plate making QR code and serial number identifier identification method |
| CN106778761A (en) * | 2016-12-23 | 2017-05-31 | 潘敏 | A kind of processing method of vehicle transaction invoice |
| CN108021842A (en) * | 2017-12-28 | 2018-05-11 | 广东蜂助手网络技术股份有限公司 | A kind of optimized treatment method of Quick Response Code |
| JP2018092644A (en) * | 2018-01-25 | 2018-06-14 | ヤマト運輸株式会社 | Baggage delivery method by anonymous delivery |
| CN108492327A (en) * | 2018-02-07 | 2018-09-04 | 广州视源电子科技股份有限公司 | AOI image matching methods, system, readable storage medium storing program for executing and smart machine |
-
2018
- 2018-09-17 CN CN201811081051.6A patent/CN109359497A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102254144A (en) * | 2011-07-12 | 2011-11-23 | 四川大学 | Robust method for extracting two-dimensional code area in image |
| CN102693409A (en) * | 2012-05-18 | 2012-09-26 | 四川大学 | Method for quickly identifying two-dimension code system type in images |
| CN103500323A (en) * | 2013-09-18 | 2014-01-08 | 西安理工大学 | Template matching method based on self-adaptive gray-scale image filtering |
| CN105976001A (en) * | 2016-05-19 | 2016-09-28 | 杭州博安实业有限公司 | False-proofing license plate making QR code and serial number identifier identification method |
| CN106778761A (en) * | 2016-12-23 | 2017-05-31 | 潘敏 | A kind of processing method of vehicle transaction invoice |
| CN108021842A (en) * | 2017-12-28 | 2018-05-11 | 广东蜂助手网络技术股份有限公司 | A kind of optimized treatment method of Quick Response Code |
| JP2018092644A (en) * | 2018-01-25 | 2018-06-14 | ヤマト運輸株式会社 | Baggage delivery method by anonymous delivery |
| CN108492327A (en) * | 2018-02-07 | 2018-09-04 | 广州视源电子科技股份有限公司 | AOI image matching methods, system, readable storage medium storing program for executing and smart machine |
Non-Patent Citations (2)
| Title |
|---|
| DONG HAN等: "《2D barcode image binarization based on wavelet analysis and Otsu"s method》", 《2010 INTERNATIONAL CONFERENCE ON COMPUTER APPLICATION AND SYSTEM MODELING》 * |
| 龙清清等: "《基于最大连通区域的QR码提取和校正方法》", 《计算机工程与应用》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110222547A (en) * | 2019-04-29 | 2019-09-10 | 北京三快在线科技有限公司 | Bar code recognition and device |
| CN110222547B (en) * | 2019-04-29 | 2020-11-03 | 北京三快在线科技有限公司 | Bar code identification method and device |
| CN113643274A (en) * | 2021-08-24 | 2021-11-12 | 凌云光技术股份有限公司 | Method and device for screening two-dimensional code candidate area |
| CN114092945A (en) * | 2022-01-21 | 2022-02-25 | 广州文石信息科技有限公司 | Handwriting input method and device based on ink screen |
| CN114092945B (en) * | 2022-01-21 | 2022-04-19 | 广州文石信息科技有限公司 | Handwriting input method and device based on ink screen |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9898808B1 (en) | Systems and methods for removing defects from images | |
| US11657631B2 (en) | Scalable, flexible and robust template-based data extraction pipeline | |
| CN110569878B (en) | Photograph background similarity clustering method based on convolutional neural network and computer | |
| US11106891B2 (en) | Automated signature extraction and verification | |
| CN110222687B (en) | Complex background card surface information identification method and system | |
| CN109359497A (en) | A kind of positioning and recognition methods in VAT invoice two-dimensional image code region | |
| CN110119741A (en) | A kind of card card image information recognition methods having powerful connections | |
| CN110569769A (en) | image recognition method and device, computer equipment and storage medium | |
| CN108280829A (en) | Welding seam method, computer installation and computer readable storage medium | |
| CN110097059B (en) | Document image binarization method, system and device based on generation countermeasure network | |
| CN110210488A (en) | The recognition methods of bar code and cell-phone number and device on a kind of express waybill | |
| Sidhwa et al. | Text extraction from bills and invoices | |
| CN105335760A (en) | Image number character recognition method | |
| CN113592735A (en) | Text page image restoration method and system, electronic equipment and computer readable medium | |
| CN109741273A (en) | A kind of mobile phone photograph low-quality images automatically process and methods of marking | |
| CN113780087A (en) | Postal parcel text detection method and equipment based on deep learning | |
| CN106384071A (en) | Two-dimensional code scanning recognition processing method and device | |
| Dergachov et al. | Data pre-processing to increase the quality of optical text recognition systems | |
| CN110414517B (en) | Rapid high-precision identity card text recognition algorithm used for being matched with photographing scene | |
| CN111055021A (en) | Visual positioning method of laser marking software | |
| Bulatov et al. | Towards a unified framework for identity documents analysis and recognition | |
| CN110942063A (en) | Certificate text information acquisition method and device and electronic equipment | |
| Rahmat et al. | Android-based text recognition on receipt bill for tax sampling system | |
| CN111445402B (en) | Image denoising method and device | |
| Katona et al. | Distance transform and template matching based methods for localization of barcodes and QR codes |
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: 20190219 |
|
| RJ01 | Rejection of invention patent application after publication |