CN105678612A - Mobile terminal original certificate electronic intelligent filling system and method - Google Patents
Mobile terminal original certificate electronic intelligent filling system and method Download PDFInfo
- Publication number
- CN105678612A CN105678612A CN201511029446.8A CN201511029446A CN105678612A CN 105678612 A CN105678612 A CN 105678612A CN 201511029446 A CN201511029446 A CN 201511029446A CN 105678612 A CN105678612 A CN 105678612A
- Authority
- CN
- China
- Prior art keywords
- image
- voucher
- module
- mobile terminal
- original certificate
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q40/00—Finance; Insurance; Tax strategies; Processing of corporate or income taxes
- G06Q40/12—Accounting
- G06Q40/125—Finance or payroll
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Finance (AREA)
- Accounting & Taxation (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Development Economics (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Technology Law (AREA)
- General Business, Economics & Management (AREA)
- Character Discrimination (AREA)
- Character Input (AREA)
Abstract
The present invention provides a mobile terminal original certificate electronic intelligent filling system and method. The method comprises a digital image acquisition step of transforming an original certificate into a digital image; an image identification processing step of carrying out the image identification processing on a to-be-identified image; and a result output step of deriving an image identification result to a financial information system. The image identification processing step comprises an image pre-processing step, a certificate type classification step, a certificate layout analysis step, a character identification step and an identification result checking step, a certificate classifier based on a decision tree is used in the certificate type classification step, an elastic template based on a hypothesis tree is used in the certificate layout analysis step, and the identification result check based on rules is used in the identification result checking step. The system applies the above method. According to the present invention, the certificate data in the certificate document images is obtained automatically, the certificate data is transformed into the corresponding business data needing to be recorded in a business system, and the business data recording efficiency is improved.
Description
Technical field
The invention belongs to intelligent identifying system field, it is specifically related to a kind of mobile terminal original certificate electronization intelligence and fills out monophyly and method.
Background technology
When submitting an expense account on the net, it is necessary to by the Data Enter in original certificate in financial information system. Traditional electronic pattern adopts the mode of manual typing to realize, reimbursement personnel are by dialog box, drop-down frame or every information of mode typing original certificate one by one of directly inputting, not only process is loaded down with trivial details, and exactness also can not be guaranteed, under traditional data-entry-form, approver cannot see the image of papery voucher, it is therefore desirable to by scanner, first-class terminating unit of making a video recording by the original certificate digitizing of papery, then annex as electronics expense report is had access in order to approver. Electronic certificate, according to pre-configured procedure information, is submitted to approver together with original certificate image and is examined by system of simultaneously submitting an expense account. Current main flow financial software is all generally exist with the form of annex for original certificate collection, an expense report association a lot of original certificate digitized images through collection. These digital pictures are generally by the time order and function order arrangement gathered, associated viscera on original certificate does not carry out identifying arrangement, cannot carrying out sequencing selection according to amount of money size, content, date etc. when having access to, this reduces the efficiency of on-line approval to a certain extent.
Applying voucher intelligent identification technology at present to be the certificate image industry of bank the most widely, mature system has electronics stamp checking system and voucher in national Check image exchange system, complete or collected works to carry back centralized processing system. Wherein the whole nation Check image exchange system use image technology material object check is converted to image and electronics clearance information, by computer techno-stress by check image and electronics clearance information be passed to maker open an account bank prompts pay the bill; In complete or collected works, electronics stamp checking system is mainly used in the voucher seal verification business of bank outlets' sales counter, adopt image procossing and mode identification technology, by the voucher seal image collected with image capture device (scanner, cleaning-sorting machine) and reserved seal image are carried out mechanical check, it is achieved the automatic business processing of seal verification;Voucher is carried back centralized processing system and is utilized the image capture device such as cleaning-sorting machine, high speed scanner batch to obtain certificate image, utilizes image processing and recognition technology to complete voucher service set in streamline operration mode and handles; In addition, the image acquisition mode of current voucher recognition system is scanner, bulky, not easily carry, cannot as the acquisition mode of mobile terminal. Therefore, it is necessary that developing an original certificate in mobile terminal identifies input system automatically, there is provided automation services can not only to reimbursement personnel, promote user experience, voucher image can also be provided to approving person, as network audit foundation, more ensure that the cognation of voucher image with every business in online reimbursement flow process.
Summary of the invention
The main purpose of the present invention is to provide a kind of mobile terminal original certificate electronization intelligence and fills out folk prescription method.
It is a further object of the present invention to provide a kind of mobile terminal original certificate electronization intelligence and fill out monophyly.
For realizing above-mentioned main purpose, original certificate electronization intelligence in mobile terminal provided by the invention is filled out folk prescription method and is comprised, digital image acquisition step: original certificate is converted to digital picture; Image recognition processing step: treat recognition image and carry out image recognition processing; Result exports step: image recognition result exports to financial information system; Wherein, the step of image recognition processing comprises Image semantic classification step, voucher type classification step, voucher space of a whole page analytical procedure, character recognition step and recognition result verification step successively; Voucher type classification step adopts the voucher sorter based on decision tree; Voucher space of a whole page analytical procedure adopts the Elastic forming board based on hypothesis tree; Recognition result is verified in step and is adopted the recognition result based on rule to verify.
From such scheme, after the identifying processing of the electronic certificate image that digital image acquisition apparatus is gathered, automatically obtain the Credential data in voucher document image, and be converted to the business datum that to need to be entered in business system corresponding, help user to improve business datum typing work.
A preferred scheme is, digital image acquisition step carries out image collection by carrying the mobile terminal of camera.
Therefore, utilize the mobile terminal carrying camera to do image collection, thus image capture device is easily obtained and easy to carry, be conducive to mobile office.
A preferred scheme is, Image semantic classification step comprises use Hough transform corrected perspective distortion algorithms or figure is carried out image recognition processing by Homomorphic Filtering Algorithm.
Therefore, due to the picture quality of mobile equipment collection compare scanner, fast instrument of clapping decrease, it is thus desirable to take preprocessing means to be optimized. Except traditional bill images pre-treatment is as except going frame line, correction, denoising, two values, also added Hough transform corrected perspective distortion algorithms or Homomorphic Filtering Algorithm, to improve picture quality.
A preferred scheme is, recognition result verify step comprise grouping verify, dictionary verify or based on contextual verification.
Therefore, recognition result is verified in step and is adopted the verification based on rule, wherein based on rule verification mainly comprise grouping verify, dictionary verify or based on contextual verification, after system identification goes out single character, need to pass through aftertreatment, utilize context information, grammer and logic that recognition result is carried out further correction, thus improve the overall performance of system. For example, when check is carried out amount of money identification, due to there is check under normal circumstances large and small to write the amount of money mutually corresponding, so identify respectively large and small write the amount of money after, so that it may with by this two portions recognition result is compared, mutual correction result.
In order to realize another object of the present invention, original certificate electronization intelligence in mobile terminal provided by the invention is filled out monophyly and is comprised, and digital image capturing module, for being converted to digital picture original certificate;Image recognition processing module, for carrying out image recognition processing image to be identified; Result output module, for exporting to financial information system by the result that image recognition processing module exports; Image recognition processing module comprises image pre-processing module, voucher type classification module, voucher space of a whole page analysis module, character recognition module and recognition result and verifies module; Voucher type classification module adopts the voucher sorter based on decision tree; The voucher space of a whole page is analyzed in module and is adopted the Elastic forming board based on hypothesis tree; Recognition result is verified in module and is adopted the recognition result based on rule to verify.
From such scheme, with based on technology such as digital image processing, voucher type classification, the analysis of the voucher space of a whole page, OCR optical character recognition and pattern recognitions, enterprise operation system is helped automatically to process voucher bills data, reach the automatic acquisition corresponding business datum of business system from batch original certificate information data, it is to increase the working efficiency of original certificate document typing.
Accompanying drawing explanation
Fig. 1 is the distributed structure block diagram that mobile terminal of the present invention original certificate electronization intelligence fills out monophyly embodiment.
Fig. 2 is the business model figure that mobile terminal of the present invention original certificate electronization intelligence fills out monophyly embodiment.
Fig. 3 is the service end processing flow chart that mobile terminal of the present invention original certificate electronization intelligence fills out single embodiment of the method.
Fig. 4 be mobile terminal of the present invention original certificate electronization intelligence fill out single embodiment of the method call identification service flow diagram.
Fig. 5 is the operation stage schema that mobile terminal of the present invention original certificate electronization intelligence fills out single embodiment of the method.
Fig. 6 is the method flow diagram that mobile terminal of the present invention original certificate electronization intelligence fills out the voucher type classification of single embodiment of the method.
Below in conjunction with drawings and Examples, the invention will be further described.
Embodiment
See Fig. 1, in actual applications, for the great amount of images data of multi-user, need to adopt distributed processing mode, this is embodied in distributed image collection and Distributed identification two aspects: multi-user initiates voucher identification request by financial information system 1, and uploads with mobile phone collection and pretreated voucher image. Available service device network request being distributed in a server cluster 3 by load balancing device 2 is got on, and image uploads to OCR Optical Character Recognition system 4 again. Image file is kept in the file system of server by OCR Optical Character Recognition system 4, and recognition result feeds back to financial information system 1, and user's examination & verification, confirmation recognition result, finally complete the automatic input of data.
See Fig. 2, Fig. 2 is that mobile terminal voucher electronization intelligence fills out single business model figure, concrete business procedure is: reimbursement people enables mobile terminal application, first steps A 1 is performed by financial information platform front end request original certificate Intelligent Recognition, perform steps A 2 system display voucher templates again to be directed at voucher image for reimbursement people and then perform steps A 3 and carry out image procossing, then perform steps A 4 and original certificate image is carried out pre-treatment, after compression, upload images information is saved to the image file system of financial information system service end, the service of steps A 5 financial information system service end application start OCR optical character recognition is performed after execution of step A4, then perform steps A 6 to export and the text information that returns on original certificate, namely financial information is sent to the financial information system front end of mobile terminal application, perform the application of steps A 7 mobile terminal afterwards to be revised for reimbursement people by financial information system front end display financial information, and the information after confirming is saved in the financial information database of service end application.Financial approval people enables mobile terminal application, perform steps A 8 and send request information to financial information system front end, then the financial information database that steps A 9 is applied is performed by financial information system front end access services end, owing between financial information database with image file system being corresponding relation, the voucher image of every reimbursement business and correspondence thereof can be checked, finally draw auditing result.
See Fig. 3 composition graphs 4, service end treatment scheme is, first perform step S1 and download image, then perform step S2 and call voucher identification service, the technological difficulties that whole mobile terminal voucher electronization intelligence fills out single business model are that the voucher identification of service end is served, step S2 is divided into again multiple step, first the voucher image that step S21 receives mobile terminal and imports into is performed, then perform step S22 to carry out image separating pressure, pre-treatment, execution step S23 calls sorter and different classes of voucher is included into respective folder (such as plane ticket afterwards, train ticket, food and drink invoice, lodging invoice, taxi ticket, VAT invoice etc.), then perform step S24 and judge whether voucher classifies successfully, if then performing step S26, otherwise perform step S25 not process. wherein the successful often kind of voucher classified is called exclusive template to carry out space of a whole page analysis by step S26, and then the word on voucher is carried out OCR optical character recognition, finally perform step S27 and recognition result is carried out result verification, and export the result file of .XML form. and then perform step S3 after executing above-mentioned steps S2 and carry out analyzing XML by machine learning, i.e. analysis result file, so that it may to perform step S4, return results and identification word is filled into corresponding position, reach and automatically fill out single object.
See the operation stage schema that Fig. 5, Fig. 5 are the present invention. Step S101 is image acquisition phase, and it is obtain the hardware device of voucher image and system that associated control software forms that image acquisition part is divided. For meeting the needs of voucher pattern recognition particularly seal authenticity identification, being that binary map picture must reach more than 200dpi precision to the basic demand gathering voucher image, 256 grades of gray-scale map pictures and 24 coloured images must reach the precision of more than 150dpi.
The camera that the present invention uses mobile terminal to carry does image collection, and non-traditional scanner, fast instrument equipment of clapping, the advantage done like this be image capture device easily obtain, easy to carry, be conducive to mobile office. But from the picture quality gathered, mobile phone and scanner, high clap instrument to compare inferior position also very obvious, just owing to mobile phone may introduce the reason such as noise jamming and voucher self-pollution in image acquisition process, image is occurred, and noise even occurs degenerating, for convenience of follow-up analyzing and processing, often need the original certificate image collected is carried out pre-treatment to improve picture quality.
Step S102 is image pre-processing phase, and image pre-processing phase is adjusted by identification area image, is become the process of the data that can carry out feature extraction. At image pre-processing phase, then result images is carried out filtering and denoising operation by the pre-print information such as the frame line first needing to remove in voucher image, then carry out image binaryzation process according to identification requirement, the character in image is extracted from complex background. Common interference has frame line, end line and the seal etc. that random noise, form ruling, specification user fill in.In addition, owing to the follow-up character recognition stage often uses individual character sorter, therefore also to be split extracting character string image, and the monocase cut out is carried out reprocessing, as reforming of monocase is processed.
According to feature extraction method and to the different requirements of the aspects such as system real time, the algorithm of Image semantic classification and process are incomplete same in each OCR Optical Character Recognition system. For the voucher image of camera collection, Image semantic classification mainly solves following image problem:
The first, Hough transform corrected perspective distortion. Owing to the arbitrariness of image angle taken by digital camera, imaging plane is difficult to completely parallel with original image place plane, in various degree flexible making that originally smooth vertical and horizontal line of text produces in a different direction, original histogram picture becomes irregular tetragon, and this kind of distortion is called as perspective distortion. Tradition adopts the method for Hough transform that OCR optical character recognition image is carried out slant correction. But, traditional method is mainly for scan image, and therefore slant correction only completes on two dimensional planes, seldom relates to three-dimensional space. The image-forming principle of handset image can use pinhole imaging system to carry out analogy, and the formula using perspective conversion is described. Like this, by the research method and projection geometry that convert will be had an X-rayed as mathematics instrument, the perspective distortion correction of voucher image is carried out.
2nd, homographic filtering. Homographic filtering, for eliminating uneven illumination, owing to the reason in incident light direction causes uneven illumination, causes carrying out image threshold segmentation (two values) inaccurate, and some critical words of the position of illumination deficiency have been divided into background, thus cause information dropout. Homographic filtering is a kind of method of frequency domain filtering, for the multiplicative noise (illumination is not enough) in removal of images, therefore may be used for solving the problem of the voucher image irradiation inequality caused by incident light direction. The brightness scope of homographic filtering one side compressed image, while strengthening the contrast gradient of image, its thinking is: by being taken the logarithm by the multiplicative noise in image, is converted into adding property noise and processes in a frequency domain, and then its inverse transformation returned by exponent arithmetic. The treating processes of homographic filtering can represent:
Wherein FFT and (FFT)-1Being respectively fourier transformation and Fourier's inverse transformation, be the transport function in frequency domain, its functional form is relevant with the object of image enhaucament.
3rd, filter background colour. The identification interference problem run into for red fire in a stove before fuel is added ticket is generally the line of the red end of bill, and blue seal during ticket checking, and this problem causes the weak effect of image binaryzation. In order to solve the problem, two kinds of pretreatment processs are attempted, direct thresholding method, by the direct filtering of pixel of RGB channel low lightness; And gray scale superposition method, process by the R passage of original color image and the gray-scale value phase Calais of channel B, by the observation of image and test, the process effect of two kinds of methods is all better than traditional method.
4th, remove seal. Seal appears on most of original certificate, and frequent generation seal covers the situation that word causes OCR optical character recognition mistake simultaneously, it is thus desirable to take image processing means to remove seal. The present invention adopts the mode directly getting Color Channel to filter the color corresponding to seal to reach the effect removing seal. Traditional method not removing only red chapter, and also the red word on voucher also together filtering, this is unfavorable for OCR optical character recognition below.Other parts on voucher are not had an impact by the red chapter of method elimination provided by the invention, and successful is better than traditional method.
Step S103 is the voucher type classification stage, and voucher classification refers to the image for multiple original certificate, sorts out according to features such as its structure, image, words, further, also need to identify the industry at voucher place, such as lodging class, food and drink class etc., to complete automatically making a report on of expense report. According to automatically filling out single business demand, user inputs multiple different classes of voucher images, and we need it to be done automatic classification on backstage, to mate respective template, and a final formation complete reimbursement document. Composition graphs 6, Fig. 6 is the method flow diagram of voucher type classification, first perform step S201 and obtain a voucher image, then perform step S202 and judge whether voucher image mates certain fixed sturcture template, if then performing step S206 to extract data, otherwise perform step S203, whether step S203 is successfully classified voucher image by voucher type classification device for judging, if then performing step S205 to mate, then perform step S206 and extract data, otherwise perform step S204 and all templates are all traveled through one time, then perform step S206 and extract data.
Take credential characteristics as foundation, design a voucher sorter based on decision Tree algorithms. The decision tree that the present invention refers to is exactly a tree structure, each node represents class bill and a feature thereof, subtree represents all subclasses and the feature thereof of this kind of bill, such as: the subclass of travel invoice comprises air ticket, steamer ticket, high guaranteed votes, bus ticket, subway ticket etc. Decision tree is the sorting technique based on rule, the similarity of the weight computing bill feature on trunk, realizes bill by searching maximum path and automatically classifies. Utilize this sorter to a certain template, and can not need all to travel through all templates one time by quick position when the space of a whole page of voucher identification is analyzed, therefore can greatly save the time.
Step S104 is the voucher space of a whole page analysis phase, space of a whole page analysis is the basis carrying out template recognition, mainly analyze the logical organization of voucher image to be identified, differentiate the kind of voucher, then by the credential information of the result of type identification and template record, analyze the accurate position of space of a whole page element, locate from original image and extract separate region to be identified subgraph. Space of a whole page analysis generally adopts the mode of template to carry out, and is every class voucher design template, applies mechanically template and is separated by the element (as the side of purchasing, the name of an article, amount of money etc.) in voucher, and then carries out OCR optical character recognition. Template can be divided into two big classes: fixing (structurizing) template and elasticity (non-structure) template, and fixed die plate is applicable to structure to be fixed and the geostationary voucher of element position. But, the structure of some voucher is not fixed, and now fixed die plate is not suitable for, it is possible to by adopting the Elastic forming board based on hypothesis tree to solve this type of problem.
Elastic forming board based on hypothesis tree is mainly used in, before voucher being OCR and identifies, it is necessary to utilize template to carry out the space of a whole page analysis of voucher, to determine to identify the concrete implication of character. Traditional fixed die plate is only applicable to that structure is fixed, the geostationary voucher of element position, and its range of application is narrow, is not suitable for the voucher image that mobile equipment gathers. Therefore, based on semantic and the rule Elastic forming board that has been every class voucher design, it may also be useful to great amount of samples trains template to analyze voucher structure.
During the space of a whole page of voucher is analyzed, the basic thought of voucher category identification is matching judgment. Different business generally has the voucher of different-format, by the analysis to the pre-print information in image, extract the various features representing unknown form, mate one by one with the voucher templates prestored, use certain criterion to differentiate, from standard template base, find out the stereotype closest to form to be identified as recognition result according to the result of similarity mode. Due to factors such as the relative position of complicated various, the space of a whole page element of voucher form and title or frame line are unstable, picture quality alternates betwwen good and bad so that the layout structure of extraction is not accurate enough, the precision that impact identifies. Thus in the space of a whole page is analyzed, usually use local feature location, after dividing out region to be identified, in subrange, carry out characteristic matching, greatly strengthen the adaptability that the voucher space of a whole page is analyzed. For the result that the space of a whole page is analyzed, follow-up to be carried out be exactly template recognition work, space of a whole page analysis can be regarded as the pre-treatment process of voucher automated processing system by us, the treating processes of system after space of a whole page analysis is summed up the recognition process into the space of a whole page, figure in the space of a whole page, graphic information and structural relation are identified and understands, it is achieved to the identification filling in territory (such as hand-written character).
Step S105 is the OCR optical character recognition stage, as the core of voucher identification, voucher image is after extracting through identification territory, and character string generally includes multiple character, need to be cut into further and for the word character identified, validity feature can be extracted and utilize sorter to realize identifying. The OCR optical character recognition stage includes feature extraction and classifier design, and character feature is generally divided into statistical nature and constitutional features according to its mode generated. According to the viewpoint of statistics, good feature extraction method must meet following 3 points: the feature of extraction is separate or uncorrelated; Inter-object distance can be effectively reduced, increase class spacing; The dimension of proper vector is as far as possible little.
The effect of sorter is the classification utilizing the syntax rule obtained in advance or decision function to differentiate character to be identified, and this process obtaining syntax rule or decision function is just called training or study. The sorting algorithm of character recognition has a variety of, and sorter conventional at present can be divided into template matches sorter, statistics decision-making sorter, syntax structure classifier, fuzzy judgment sorter, neural network classifier and reasoning from logic sorter etc. In actual applications, the method for multiple Classifiers Combination is also often used.
Step S106 be result verify the stage, result verify in adopt the verification based on rule, wherein based on rule verification mainly comprise grouping verify, dictionary verify, based on contextual verification. After system identification goes out single character, it is necessary to by aftertreatment, utilize context information, grammer and logic that recognition result is carried out further correction, to improve the overall performance of system. For example, when check is carried out amount of money identification, due to there is check under normal circumstances large and small to write the amount of money mutually corresponding, so identify respectively large and small write the amount of money after, so that it may with by this two portions recognition result is compared, mutual correction result.
The present invention can realize automatically classifying from the digital picture of one batch of service application associated documents document, then corresponding business datum in the rear voucher document image of classification is automatically obtained, and business datum typing the present invention filled out in monophyly, can also by the view data of the business datum real-time tracing of automatic input to extracted information, audit so that the business datum automatically obtained is carried out verification according to voucher image by supplementary typing personnel, it is to increase working efficiency.
Last it is emphasized that above-described embodiment is only the preferred scheme of the present invention, can also there is more change during practical application, such as, carry the use of the different mobile terminal of camera; Or, recognition result is verified in module and is used different check methods, and such change also can realize the object of the present invention, also should be included in the protection domain of the claims in the present invention.
Claims (8)
1. mobile terminal original certificate electronization intelligence fills out folk prescription method, comprises
Digital image acquisition step: original certificate is converted to digital picture;
Image recognition processing step: treat recognition image and carry out image recognition processing;
Result exports step: image recognition result exports to financial information system;
It is characterized in that:
Described image recognition processing step comprises Image semantic classification step, voucher type classification step, voucher space of a whole page analytical procedure, character recognition step and recognition result successively and verifies step;
Described voucher type classification step adopts the voucher sorter based on decision tree;
Described voucher space of a whole page analytical procedure adopts the Elastic forming board based on hypothesis tree;
Described recognition result is verified in step and is adopted the recognition result based on rule to verify.
2. mobile terminal according to claim 1 original certificate electronization intelligence fills out folk prescription method, it is characterised in that:
Described digital image acquisition step carries out image collection by carrying the mobile terminal of camera.
3. mobile terminal according to claim 1 original certificate electronization intelligence fills out folk prescription method, it is characterised in that:
Described Image semantic classification step use Hough transform corrected perspective distortion algorithms or Homomorphic Filtering Algorithm image is carried out image recognition processing.
4. mobile terminal according to claim 1 original certificate electronization intelligence fills out folk prescription method, it is characterised in that:
Described recognition result verify step comprise grouping verify, dictionary verify or based on contextual verification.
5. mobile terminal original certificate electronization intelligence fills out monophyly, comprises
Digital image capturing module, for being converted to digital picture original certificate;
Image recognition processing module, for carrying out image recognition processing image to be identified;
Result output module, for exporting to financial information system by the result that image recognition processing module exports;
It is characterized in that:
Described image recognition processing module comprises image pre-processing module, voucher type classification module, voucher space of a whole page analysis module, character recognition module and recognition result and verifies module;
Described voucher type classification module adopts the voucher sorter based on decision tree;
The described voucher space of a whole page is analyzed in module and is adopted the Elastic forming board based on hypothesis tree;
Described recognition result is verified in module and is adopted the recognition result based on rule to verify.
6. mobile terminal according to claim 5 original certificate electronization intelligence fills out monophyly, it is characterised in that:
Described digital image capturing module is the mobile terminal carrying camera.
7. mobile terminal according to claim 5 original certificate electronization intelligence fills out monophyly, it is characterised in that:
Described image pre-processing module comprises of Hough transform corrected perspective distortion algorithms module or Homomorphic Filtering Algorithm module.
8. mobile terminal according to claim 5 original certificate electronization intelligence fills out monophyly, it is characterised in that:
Described recognition result is verified module and is comprised that module is verified in grouping, module verified by dictionary or based on of contextual verification module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511029446.8A CN105678612A (en) | 2015-12-30 | 2015-12-30 | Mobile terminal original certificate electronic intelligent filling system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511029446.8A CN105678612A (en) | 2015-12-30 | 2015-12-30 | Mobile terminal original certificate electronic intelligent filling system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105678612A true CN105678612A (en) | 2016-06-15 |
Family
ID=56189925
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511029446.8A Pending CN105678612A (en) | 2015-12-30 | 2015-12-30 | Mobile terminal original certificate electronic intelligent filling system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105678612A (en) |
Cited By (41)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106504079A (en) * | 2016-09-22 | 2017-03-15 | 江苏富山企业服务有限公司 | A kind of composite type financial management method and its management platform |
| CN106557747A (en) * | 2016-11-15 | 2017-04-05 | 平安科技（深圳）有限公司 | The method and device of identification insurance single numbers |
| CN106569059A (en) * | 2016-11-01 | 2017-04-19 | 广西电网有限责任公司电力科学研究院 | Production service system with function of converting test data to structured storage |
| CN106650718A (en) * | 2016-12-21 | 2017-05-10 | 远光软件股份有限公司 | Certificate image identification method and apparatus |
| CN107256515A (en) * | 2017-07-04 | 2017-10-17 | 深圳易嘉恩科技有限公司 | The method of the financial integrated OCR identification softwares of cloud platform |
| CN107392221A (en) * | 2017-06-05 | 2017-11-24 | 天方创新（北京）信息技术有限公司 | The method and device of the training method of disaggregated model, OCR recognition results of classifying |
| CN107657526A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank is convenient to fill out single device |
| CN107790403A (en) * | 2017-10-18 | 2018-03-13 | 四川长虹电器股份有限公司 | A kind of sorting system of Financial Billing and the method for sorting of Financial Billing |
| CN107886422A (en) * | 2017-10-19 | 2018-04-06 | 远光软件股份有限公司 | One kind reimbursement equipment and its processing method |
| CN108346106A (en) * | 2018-02-23 | 2018-07-31 | 平安科技（深圳）有限公司 | Bill entry method, system, optical character recognition server and storage medium |
| CN108363943A (en) * | 2017-12-27 | 2018-08-03 | 苏州工业园区报关有限公司 | Clearance robot based on Weigh sensor technology |
| CN108416895A (en) * | 2018-03-16 | 2018-08-17 | 四川长虹电器股份有限公司 | A kind of enterprise's invoice input system and method based on image recognition technology |
| CN109064304A (en) * | 2018-08-03 | 2018-12-21 | 四川长虹电器股份有限公司 | Finance reimbursement bill automated processing system and method |
| WO2018233171A1 (en) * | 2017-06-23 | 2018-12-27 | 平安科技（深圳）有限公司 | Method and apparatus for entering document information, computer device and storage medium |
| CN109145760A (en) * | 2018-07-27 | 2019-01-04 | 苏州浪潮智能软件有限公司 | Intelligence fills out single method, apparatus, computer equipment and storage medium |
| CN109190629A (en) * | 2018-08-28 | 2019-01-11 | 传化智联股份有限公司 | A kind of electronics waybill generation method and device |
| CN109345366A (en) * | 2018-09-17 | 2019-02-15 | 程浩 | A kind of financial affairs receipt management system and method |
| CN109447002A (en) * | 2018-10-31 | 2019-03-08 | 广州慧睿思通信息科技有限公司 | Government affairs business material preliminary examination method, apparatus, equipment and storage medium |
| CN109446345A (en) * | 2018-09-26 | 2019-03-08 | 深圳中广核工程设计有限公司 | Nuclear power file verification processing method and system |
| CN109767048A (en) * | 2017-11-08 | 2019-05-17 | 中国石油天然气股份有限公司 | Assets collecting method and method for processing resource |
| CN109767545A (en) * | 2017-01-10 | 2019-05-17 | 中国人民银行印制科学技术研究所 | The defect classification method and defect categorizing system of valuable bills |
| CN109840519A (en) * | 2019-01-25 | 2019-06-04 | 青岛盈智科技有限公司 | A kind of adaptive intelligent form recognition input device and its application method |
| CN109919125A (en) * | 2019-03-19 | 2019-06-21 | 厦门商集网络科技有限责任公司 | Travel stroke restoring method and its system based on bank slip recognition |
| CN110009075A (en) * | 2019-01-28 | 2019-07-12 | 秒针信息技术有限公司 | Processing method, device, storage medium and the electronic device of target information |
| CN110209632A (en) * | 2019-05-27 | 2019-09-06 | 武汉市润普网络科技有限公司 | A kind of electronics folder with case production, turn shelves system |
| CN110443317A (en) * | 2019-08-09 | 2019-11-12 | 上海尧眸电气科技有限公司 | A kind of method, apparatus and electronic equipment of paper shelves electronic data processing |
| CN110490181A (en) * | 2019-08-14 | 2019-11-22 | 北京思图场景数据科技服务有限公司 | A kind of list based on OCR identification technology fills in checking method, device, equipment and computer storage medium |
| CN110599317A (en) * | 2019-08-26 | 2019-12-20 | 湖南大唐先一科技有限公司 | Account reporting and auditing automation method based on rule engine and OCR (optical character recognition) |
| CN110619252A (en) * | 2018-06-19 | 2019-12-27 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for identifying form data in picture and storage medium |
| CN110909733A (en) * | 2019-10-28 | 2020-03-24 | 世纪保众(北京)网络科技有限公司 | Template positioning method and device based on OCR picture recognition and computer equipment |
| CN110929580A (en) * | 2019-10-25 | 2020-03-27 | 北京译图智讯科技有限公司 | Financial statement information rapid extraction method and system based on OCR |
| CN111079755A (en) * | 2019-07-26 | 2020-04-28 | 中央军委后勤保障部财务局 | Financial reimbursement data processing method, device and system |
| CN111242760A (en) * | 2019-12-30 | 2020-06-05 | 航天信息股份有限公司企业服务分公司 | Method and system for carrying out accounting on capital service based on capital institution |
| CN111311197A (en) * | 2020-03-05 | 2020-06-19 | 中国工商银行股份有限公司 | Travel data processing method and device |
| CN111767818A (en) * | 2020-06-23 | 2020-10-13 | 北京思特奇信息技术股份有限公司 | Method and device for intelligently accepting service |
| CN113095307A (en) * | 2021-06-09 | 2021-07-09 | 国网浙江省电力有限公司 | Automatic identification method for financial voucher information |
| CN113130023A (en) * | 2021-04-22 | 2021-07-16 | 嘉兴易迪希计算机技术有限公司 | Image-text recognition and entry method and system in EDC system |
| CN113344096A (en) * | 2021-06-22 | 2021-09-03 | 郑州信源信息技术股份有限公司 | Automatic bid document analysis method and system based on OCR technology |
| CN113449698A (en) * | 2021-08-30 | 2021-09-28 | 湖南文盾信息技术有限公司 | Automatic paper document input method, system, device and storage medium |
| CN114648776A (en) * | 2022-05-24 | 2022-06-21 | 威海海洋职业学院 | Financial reimbursement data processing method and processing system |
| CN117373030A (en) * | 2023-06-19 | 2024-01-09 | 上海简答数据科技有限公司 | OCR-based user material identification method, system, device and medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737242A (en) * | 2012-06-12 | 2012-10-17 | 丰豪盈彩（北京）科技有限公司 | Automatic bill recognition method and system applied to mobile terminal |
| CN103488999A (en) * | 2013-09-11 | 2014-01-01 | 东华大学 | Invoice data recording method |
| CN103995904A (en) * | 2014-06-13 | 2014-08-20 | 上海珉智信息科技有限公司 | Recognition system for image file electronic data |
| CN104751194A (en) * | 2015-04-27 | 2015-07-01 | 陈包容 | Processing method and processing device for financial expense reimbursement |
-
2015
- 2015-12-30 CN CN201511029446.8A patent/CN105678612A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102737242A (en) * | 2012-06-12 | 2012-10-17 | 丰豪盈彩（北京）科技有限公司 | Automatic bill recognition method and system applied to mobile terminal |
| CN103488999A (en) * | 2013-09-11 | 2014-01-01 | 东华大学 | Invoice data recording method |
| CN103995904A (en) * | 2014-06-13 | 2014-08-20 | 上海珉智信息科技有限公司 | Recognition system for image file electronic data |
| CN104751194A (en) * | 2015-04-27 | 2015-07-01 | 陈包容 | Processing method and processing device for financial expense reimbursement |
Cited By (55)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106504079A (en) * | 2016-09-22 | 2017-03-15 | 江苏富山企业服务有限公司 | A kind of composite type financial management method and its management platform |
| CN106569059A (en) * | 2016-11-01 | 2017-04-19 | 广西电网有限责任公司电力科学研究院 | Production service system with function of converting test data to structured storage |
| CN106557747A (en) * | 2016-11-15 | 2017-04-05 | 平安科技（深圳）有限公司 | The method and device of identification insurance single numbers |
| CN106557747B (en) * | 2016-11-15 | 2018-06-22 | 平安科技（深圳）有限公司 | The method and device of identification insurance single numbers |
| CN106650718A (en) * | 2016-12-21 | 2017-05-10 | 远光软件股份有限公司 | Certificate image identification method and apparatus |
| CN109767545A (en) * | 2017-01-10 | 2019-05-17 | 中国人民银行印制科学技术研究所 | The defect classification method and defect categorizing system of valuable bills |
| CN109767545B (en) * | 2017-01-10 | 2021-06-08 | 中钞印制技术研究院有限公司 | Method and system for classifying defects of valuable bills |
| CN107392221B (en) * | 2017-06-05 | 2020-09-22 | 天方创新（北京）信息技术有限公司 | Training method of classification model, and method and device for classifying OCR (optical character recognition) results |
| CN107392221A (en) * | 2017-06-05 | 2017-11-24 | 天方创新（北京）信息技术有限公司 | The method and device of the training method of disaggregated model, OCR recognition results of classifying |
| WO2018233171A1 (en) * | 2017-06-23 | 2018-12-27 | 平安科技（深圳）有限公司 | Method and apparatus for entering document information, computer device and storage medium |
| CN107256515A (en) * | 2017-07-04 | 2017-10-17 | 深圳易嘉恩科技有限公司 | The method of the financial integrated OCR identification softwares of cloud platform |
| CN107657526A (en) * | 2017-09-27 | 2018-02-02 | 安徽硕威智能科技有限公司 | A kind of bank is convenient to fill out single device |
| CN107790403B (en) * | 2017-10-18 | 2019-07-19 | 四川长虹电器股份有限公司 | A kind of sorting system of Financial Billing and the method for sorting of Financial Billing |
| CN107790403A (en) * | 2017-10-18 | 2018-03-13 | 四川长虹电器股份有限公司 | A kind of sorting system of Financial Billing and the method for sorting of Financial Billing |
| CN107886422A (en) * | 2017-10-19 | 2018-04-06 | 远光软件股份有限公司 | One kind reimbursement equipment and its processing method |
| CN109767048A (en) * | 2017-11-08 | 2019-05-17 | 中国石油天然气股份有限公司 | Assets collecting method and method for processing resource |
| CN108363943B (en) * | 2017-12-27 | 2020-12-01 | 苏州工业园区报关有限公司 | Customs clearance robot based on intelligent recognition technology |
| CN108363943A (en) * | 2017-12-27 | 2018-08-03 | 苏州工业园区报关有限公司 | Clearance robot based on Weigh sensor technology |
| WO2019161615A1 (en) * | 2018-02-23 | 2019-08-29 | 平安科技（深圳）有限公司 | Bill entry method, system, optical character recognition server and storage medium |
| CN108346106A (en) * | 2018-02-23 | 2018-07-31 | 平安科技（深圳）有限公司 | Bill entry method, system, optical character recognition server and storage medium |
| CN108416895A (en) * | 2018-03-16 | 2018-08-17 | 四川长虹电器股份有限公司 | A kind of enterprise's invoice input system and method based on image recognition technology |
| CN110619252B (en) * | 2018-06-19 | 2022-11-04 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for identifying form data in picture and storage medium |
| CN110619252A (en) * | 2018-06-19 | 2019-12-27 | 百度在线网络技术（北京）有限公司 | Method, device and equipment for identifying form data in picture and storage medium |
| CN109145760A (en) * | 2018-07-27 | 2019-01-04 | 苏州浪潮智能软件有限公司 | Intelligence fills out single method, apparatus, computer equipment and storage medium |
| CN109064304A (en) * | 2018-08-03 | 2018-12-21 | 四川长虹电器股份有限公司 | Finance reimbursement bill automated processing system and method |
| CN109190629A (en) * | 2018-08-28 | 2019-01-11 | 传化智联股份有限公司 | A kind of electronics waybill generation method and device |
| CN109345366A (en) * | 2018-09-17 | 2019-02-15 | 程浩 | A kind of financial affairs receipt management system and method |
| CN109446345A (en) * | 2018-09-26 | 2019-03-08 | 深圳中广核工程设计有限公司 | Nuclear power file verification processing method and system |
| CN109447002A (en) * | 2018-10-31 | 2019-03-08 | 广州慧睿思通信息科技有限公司 | Government affairs business material preliminary examination method, apparatus, equipment and storage medium |
| CN109840519B (en) * | 2019-01-25 | 2023-05-05 | 青岛盈智科技有限公司 | Self-adaptive intelligent bill identification and input device and application method thereof |
| CN109840519A (en) * | 2019-01-25 | 2019-06-04 | 青岛盈智科技有限公司 | A kind of adaptive intelligent form recognition input device and its application method |
| CN110009075A (en) * | 2019-01-28 | 2019-07-12 | 秒针信息技术有限公司 | Processing method, device, storage medium and the electronic device of target information |
| CN109919125A (en) * | 2019-03-19 | 2019-06-21 | 厦门商集网络科技有限责任公司 | Travel stroke restoring method and its system based on bank slip recognition |
| CN109919125B (en) * | 2019-03-19 | 2021-03-26 | 厦门商集网络科技有限责任公司 | Travel route restoration method and system based on bill recognition |
| CN110209632A (en) * | 2019-05-27 | 2019-09-06 | 武汉市润普网络科技有限公司 | A kind of electronics folder with case production, turn shelves system |
| CN111079755A (en) * | 2019-07-26 | 2020-04-28 | 中央军委后勤保障部财务局 | Financial reimbursement data processing method, device and system |
| CN111079755B (en) * | 2019-07-26 | 2021-05-25 | 中央军委后勤保障部财务局 | Financial reimbursement data processing method, device and system |
| CN110443317A (en) * | 2019-08-09 | 2019-11-12 | 上海尧眸电气科技有限公司 | A kind of method, apparatus and electronic equipment of paper shelves electronic data processing |
| CN110490181A (en) * | 2019-08-14 | 2019-11-22 | 北京思图场景数据科技服务有限公司 | A kind of list based on OCR identification technology fills in checking method, device, equipment and computer storage medium |
| CN110490181B (en) * | 2019-08-14 | 2022-04-22 | 北京思图场景数据科技服务有限公司 | Form filling and auditing method, device and equipment based on OCR (optical character recognition) technology and computer storage medium |
| CN110599317A (en) * | 2019-08-26 | 2019-12-20 | 湖南大唐先一科技有限公司 | Account reporting and auditing automation method based on rule engine and OCR (optical character recognition) |
| CN110929580A (en) * | 2019-10-25 | 2020-03-27 | 北京译图智讯科技有限公司 | Financial statement information rapid extraction method and system based on OCR |
| CN110909733A (en) * | 2019-10-28 | 2020-03-24 | 世纪保众(北京)网络科技有限公司 | Template positioning method and device based on OCR picture recognition and computer equipment |
| CN111242760B (en) * | 2019-12-30 | 2024-02-27 | 航天信息股份有限公司企业服务分公司 | Method and system for billing fund business based on fund institutions |
| CN111242760A (en) * | 2019-12-30 | 2020-06-05 | 航天信息股份有限公司企业服务分公司 | Method and system for carrying out accounting on capital service based on capital institution |
| CN111311197A (en) * | 2020-03-05 | 2020-06-19 | 中国工商银行股份有限公司 | Travel data processing method and device |
| CN111767818B (en) * | 2020-06-23 | 2024-04-26 | 北京思特奇信息技术股份有限公司 | Method and device for intelligently accepting business |
| CN111767818A (en) * | 2020-06-23 | 2020-10-13 | 北京思特奇信息技术股份有限公司 | Method and device for intelligently accepting service |
| CN113130023B (en) * | 2021-04-22 | 2023-04-07 | 嘉兴易迪希计算机技术有限公司 | Image-text recognition and entry method and system in EDC system |
| CN113130023A (en) * | 2021-04-22 | 2021-07-16 | 嘉兴易迪希计算机技术有限公司 | Image-text recognition and entry method and system in EDC system |
| CN113095307A (en) * | 2021-06-09 | 2021-07-09 | 国网浙江省电力有限公司 | Automatic identification method for financial voucher information |
| CN113344096A (en) * | 2021-06-22 | 2021-09-03 | 郑州信源信息技术股份有限公司 | Automatic bid document analysis method and system based on OCR technology |
| CN113449698A (en) * | 2021-08-30 | 2021-09-28 | 湖南文盾信息技术有限公司 | Automatic paper document input method, system, device and storage medium |
| CN114648776A (en) * | 2022-05-24 | 2022-06-21 | 威海海洋职业学院 | Financial reimbursement data processing method and processing system |
| CN117373030A (en) * | 2023-06-19 | 2024-01-09 | 上海简答数据科技有限公司 | OCR-based user material identification method, system, device and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105678612A (en) | Mobile terminal original certificate electronic intelligent filling system and method | |
| CN110414927B (en) | Method and device for automatically generating voucher during bill processing | |
| CN109543690B (en) | Method and device for extracting information | |
| Sugiarto et al. | Wood identification based on histogram of oriented gradient (HOG) feature and support vector machine (SVM) classifier | |
| CN104463195A (en) | Printing style digital recognition method based on template matching | |
| CN111968193B (en) | Text image generation method based on StackGAN (secure gas network) | |
| CN103177128A (en) | Method and system for processing bill crown word number information | |
| US10043071B1 (en) | Automated document classification | |
| CN112465596B (en) | Image information processing cloud computing platform based on electronic commerce live broadcast | |
| CN109684957A (en) | A kind of method and system showing system data according to paper form automatically | |
| Engin et al. | Offline signature verification on real-world documents | |
| Yindumathi et al. | Analysis of image classification for text extraction from bills and invoices | |
| CN114821725A (en) | Miner face recognition system based on neural network | |
| CN115062117A (en) | Method for automatically generating and classifying documents based on natural language processing technology | |
| CN111462388A (en) | Bill inspection method and device, terminal equipment and storage medium | |
| CN112508000B (en) | Method and equipment for generating OCR image recognition model training data | |
| CN110197140A (en) | Material checking method and equipment based on Text region | |
| CN104899551B (en) | A kind of form image sorting technique | |
| CN109460768B (en) | Text detection and removal method for histopathology microscopic image | |
| US11557107B2 (en) | Intelligent recognition and extraction of numerical data from non-numerical graphical representations | |
| CN116798061A (en) | Bill auditing and identifying method, device, terminal and storage medium | |
| CN111581299A (en) | Inter-library data conversion system and method of multi-source data warehouse based on big data | |
| KR102392644B1 (en) | Apparatus and method for classifying documents based on similarity | |
| CN114677333A (en) | Image contrast enhancement detection method based on histogram | |
| CN111414889A (en) | Financial statement identification method and device based on character identification |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160615 |