CN104077682A - Document data entry method based on OCR and task fragmentization - Google Patents
Document data entry method based on OCR and task fragmentization Download PDFInfo
- Publication number
- CN104077682A CN104077682A CN201410307381.8A CN201410307381A CN104077682A CN 104077682 A CN104077682 A CN 104077682A CN 201410307381 A CN201410307381 A CN 201410307381A CN 104077682 A CN104077682 A CN 104077682A
- Authority
- CN
- China
- Prior art keywords
- field
- typing
- task
- value
- ocr
- 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
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a document data entry method based on OCR and task fragmentization. The method comprises the steps that (1) document image data are read, and according to the specimen information data of a document, the image data are classified into various documents; (2) an OCR technology is used for carrying out recognition on the image data of the documents, and the content of various fields is obtained; (3) according to the OCR field content and relation setting between the various fields, whether the various fields need entry is judged; and (4) according to the OCR coordinate setting, the fields which need entry are subjected to segmentation, the fields are segmented into a plurality of fragments and are distributed into a plurality of tasks, and fragment type entry is carried out through the Internet. Then, the method can comprises the steps of data verification, field value integration, field logical check and the like. According to the method, the OCR technology and Internet resources are combined, the problems of image blur and inaccurate locating are solved, entry fields are greatly reduced, and data processing capacity, quality and efficiency can be greatly improved.
Description
Technical field
The invention belongs to image data identification and processing technology field, be specifically related to a kind of document data entry method based on OCR identification and task fragmentation.
Background technology
Generally, the treatment scheme of existing data handling system mostly:------acquiescence coordinate is set, and---finished product is exported in data typing---verification of data---to image warehouse-in in document classification.Under this operation flow, deal with data typing business, need to define project processing rule, and operating personnel is carried out data typing, checks a series of trainings of operation, and project just can formally be reached the standard grade.
Traditional data is processed the too alligatoring of system business process of company, after image warehouse-in, document classification, directly obtains acquiescence coordinate setting, middle not to image is rectified a deviation, denoising point etc. makes image definition processing.The typing task generating like this, with regard to the problem such as there will be image coordinate and typing field to depart from, image field contents is fuzzy, affects accuracy and the input speed of typing task.And owing to usining whole document as processing unit, be unfavorable for multi-person synergy operation.In addition, project quality is checked on by operating personnel completely, and system does not have a set of comprehensive logical check rule, to client's project quality, cannot obtain larger guarantee.
Existing this operation flow directly all generates typing task after getting field coordinate; centre is not identified and has been determined whether to fill in each field coordinate content of image; factor data is processed business and is often had a lot of image field contents for empty; do not carry out the direct generation task of blank judgement; will cause occurring a large amount of blank typing tasks, thereby these sleazy typing tasks can directly have influence on the risk that we operating personnel's business processing speed increase business is paid time delay.
Meanwhile, adopt current this traditional business treatment scheme, company need be equipped with a large amount of machinery and equipment, recruit a large amount of operating personnels, also needs according to business rule, to carry out a series of matters such as strong rule training, has greatly increased undoubtedly the operation cost of company.
Summary of the invention
The invention provides a kind of document data entry method based on OCR identification, task fragmentation, flow chart of data processing carried out to degree of depth refinement, and in conjunction with Internet resources, solved image fog, can not precise positioning and be difficult to the problems such as large-scale production.OCR identification has accurately reduced a large amount of typing fields, and in conjunction with the validation verification of business rule, making fully in conjunction with Internet resources, to carry out large-scale production when ensuring the quality of products becomes possibility, can greatly improve production capacity, quality and the efficiency of data processing.
For achieving the above object, the technical solution used in the present invention is as follows:
A document data entry method based on OCR identification and task fragmentation, its step comprises:
1) document classification: read the image data of document, and according to the sample information data of document, image data is divided into all kinds of document templates;
2) OCR identification: adopt OCR technology to identify the image data of document, obtain the content of each field;
3) typing policy optimization: the setting that is related to according to the field contents of OCR identification and each interfield, judges whether each field needs typing;
4) data typing:, cut into some fragments and be distributed into a plurality of typing tasks according to rule needing the field of typing to carry out cutting according to OCR coordinate setting, carry out fragment type typing by internet.
Further, step 2) described employing OCR technology is identified document, comprises image processing, field coordinate setting and field value identification, obtains the content of each field and accurate coordinate, rejects without content field simultaneously.
Further, step 4) by typing field by the difficulty of section content should, logic configuration and the proof strength of significance level, system, capable of dynamic generates needs the typing number of times carried out, by once or complete typing task for several times.
Further, step 4) carry out afterwards verification of data, and data field value is integrated.As same field, carry out the result of repeatedly typing when inconsistent, the personnel of checking can compare, revise according to the result of typing before, fill in correct field value.
Further, after integrating, field value also comprises logical check step.Logical check is that the final typing value of each field is regular according to the logical check separately configuring, and carries out logic verify and conversion, generates field and becomes performance number.Logical check is divided into: the logical check of individual character section logical check and interfield.
Further, after logical check, carry out finished product inspection and output step.
Compare with traditional data handling system, beneficial effect of the present invention is as follows:
1) the present invention carries out degree of depth refinement to flow chart of data processing, in conjunction with the Internet resources finished product of can delivering to customer more fast, in a large number.System increases the characteristics such as precise positioning (OCR recognition technology), policy optimization, logical check, internet typing, production capacity, quality and the efficiency of data processing have greatly been improved, solved image fog, can not precise positioning etc. problem, reduce a large amount of typing fields, by a large amount of business rules, carried out the correctness that logical check guarantees data simultaneously.
2) whole business processing flow of the present invention completes by configuration, and each process module is independent, can, according to service needed configuration operation flow process neatly, be enough to meet modern client's diversified demand.Business rule demand can complete by configuring directly substantially, does not need additionally to write a large amount of codes, system easy-to-use, practical.Compare the project of can reaching the standard grade faster, and the trouble-free operation in 7*24 hour that system can be highly stable with traditional data handling system.
3) the flow chart of data processing of the present invention data processing business model that breaks traditions, Business Rule Engine by the business rule of the large amount of complex for typing personnel by system backstage completes, met internet to typing fragmentation, random demand, enterprise does not need to prepare larger place, recruits a large amount of operating personnels, purchases large number quipments, carries out a large amount of numerous and diverse rule trainings, and operation cost can realize significantly reduction, business processing efficiency and can be greatly improved.
4) business process system of the present invention adopts internet equalization typing pattern, by internet platform, provide the chance of the part-time employment that spends one's leisure for masses, can accelerate to promote upgrading transition and the development of service procedure outsourcing industry, meet client's diversified demand, again can be on time when reducing self operation cost of enterprises, by matter, to client, provide better service.
Accompanying drawing explanation
Fig. 1 is the flow chart of steps of the document data entry method based on OCR identification and task fragmentation in embodiment.
Fig. 2 carries out the flow chart of steps of fragment type internet typing in embodiment.
Embodiment
Below by specific embodiments and the drawings, the present invention will be further described.
Fig. 1 is the flow chart of steps of the document data entry method based on OCR identification and task fragmentation of the present invention, and as shown in the drawing, its operation workflow is:
---document classification---OCR identification---typing policy optimization---data typing---verification of data---field value integration---field logical check---finished product inspection---output (customizing client's finished product) of image warehouse-in.
Each step in above-mentioned flow process is specifically described as follows:
1. image is put in storage
Program reads image data bag the import system of download client transmission automatically.Image data bag refers to the picture file compressed package that client forms by certain rule and the rear compression of form scanning outsourcing project image.
2. document classification
According to projects document rule of writing system, warehouse-in correction of image information is read in identification automatically, is divided into all kinds of documents corresponding with system template.System template is the sample information data of using all kinds of documents of modeling program generation.This assorting process is completed automatically by program.
3.OCR identification
OCR recognition node carries out three step processing to document: image processing, field coordinate setting, field value identification.
Image processing be to raw video rectify a deviation, sharpening processes, and makes image field contents more clear, the field location obtaining is more accurate, fast and easy typing operation.
Field coordinate setting is according to template coordinate configuration, by OCR recognition technology to needing the field of typing to carry out coordinate setting.
Field value identification is according to field coordinate position, by the fill substance of OCR technology identification field.
4. typing policy optimization
Typing policy optimization is the setting that is related to according to the field contents of OCR identification and each interfield, judges and determines whether each field needs typing.
For example: by the identification content of warrantee's name and these two fields of passport NO., judge whether whole warrantee's information (comprising sex, birthday, address, phone etc.) is empty.According to each document rule, if generally the name of customer data and passport NO. are not filled in, other information be all also sky.So whether example can be empty according to the OCR identification content of warrantee's name and two fields of passport NO., judges whether whole warrantee's information needs typing.
In addition some other single field also can judge whether to need typing according to the field identification content of OCR, as single class and the typing class field chosen of basic document, unit information, health and fitness information, financial information etc., all can determine whether to need typing according to the OCR identification content of this field: while being identified as sky, field is judged to be sky, does not need typing; While identifying meaningful and content intact, can directly get identification content as field value, also not need to carry out typing; When identification content is imperfect or None-identified goes out content, field need to generate manual entry task.
5. data typing
Typing field is become to a plurality of typing tasks by the regular allocation setting, adopt the typing of fragment type internet.The typing of fragment type internet refers to field cut into some fragments, upsets to be put into respectively online input system after order and to process.The fragment generating is that field section is complete according to the accurate cutting of OCR coordinate setting of typing field, and content is clear.Fig. 2 is the process flow diagram that carries out the typing of fragment type internet.
Adopt the typing of fragment type internet can protect preferably the safety of customer information data; fragment type typing section is meticulous brief in addition; typing personnel only need be shone figure typing; do not need to remember a large amount of typing dependency rules; each section content briefly is also conducive to improve the accuracy of typing content; utilize Internet resources can complete faster payment, shorten the task processing time.
For example: passport NO. field.Chinese document passport NO. is substantially all ID (identity number) card No., for identification card number code field, for carrying out data security work, it (can be also other quantity that system is cut into 3 fragments by its fractionation, can be according to business and field situation flexible configuration) upset and be put into respectively online input system after order and process, make two bites at a cherry.Specific practice is as follows:
1) the first two of passport NO. section field (such as first 6 and middle 8) is directly obtained previous OCR discre value, wouldn't allocating task, last field of passport NO. (last 4) is assigned to an online record (being the typing for the first time of internet);
2) after the logging data of all online typing integer fields for the first time that dispense and fractionation field is all returned, the typing value of each field be take to integer field and arrive together as unit integration, and the online typing value of the OCR value of passport NO. the first two field and last field is combined;
3) data after integrating are carried out to I.D. system check, whether see legal (checking of I.D. can adopt existing algorithm), legal next step flow process that directly enters, is assigned to the first two field of passport NO. to carry out typing again (typing for the second time of internet) on the net by secondary generation task when illegal;
4) with final typing value, carry out again the legal checking of I.D., legally directly pass through, when illegal, whole field value is recovered to internal processes and is examined by quality inspection personnel.
6. verification of data
Verification of data is two record results in same typing field when inconsistent, the task by artificial judgment input result validity of generation.Two record results are when all invalid, and the personnel of checking can revise or fill in the right value of field voluntarily.
7. field value is integrated
It is by the typing value of field and the value of checking that field value is integrated, and by the rules integration configuring, arrives together, generates the final typing value of whole part of each field of document.Integration process is completed automatically by program.
8. logical check
Logical check is that the final typing value of each field is regular according to the logical check separately configuring, and carries out logic verify and conversion, generates field and becomes performance number.Logical check is divided into the logical check of individual character section logical check and interfield.
The inspection of individual character section is carried out logic checking for single field according to configuration rule exactly.E-mail address field for example, all can there be fixing character and form in general E-mail address, such as: in E-mail address, necessarily there is a symbol etc.First configure accordingly the logical check rule of E-mail address.When typing value and rule do not meet, during through logical check flow process, will be extracted.
Inter-field check is exactly according to the relation rule between each field, relevant field to be connected and to be checked together, and when logic checking is obstructed out-of-date, system can extract certain field of relevant field or whole field according to configuration.For example: nationality and passport NO. field, passport NO. is filled in while being ID (identity number) card No., and nationality is China certainly.When nationality's typing value is not China, system will extract nationality's field or passport NO. field separately or all, again carries out the examination of typing value.
Part field can arrange the inspection of individual character section and the checking of inter-field check Dual Logic.For example: area code and postcode field, can carry out checking and verifying of individual character section according to self rule, also can combine with address field, whether area code and postcode that checking is filled in be corresponding with address information, strengthens checking on of field typing value accuracy.
9. finished product inspection
Finished product inspection is that the field that logical check authentication failed is extracted generates artificial finished product inspection task, by professional, is checked, judges and revised.After finished product inspection completes, generation is exactly that final document field becomes performance number.
10. output
Output is exactly to extract document field final finished value, according to customer demand conversion, outputs in the finished product file of corresponding format.By network service, uploading to client specifies finished product to receive catalogue.
Above embodiment is only in order to technical scheme of the present invention to be described but not be limited; those of ordinary skill in the art can modify or be equal to replacement technical scheme of the present invention; and not departing from the spirit and scope of the present invention, protection scope of the present invention should be as the criterion with described in claim.
Claims (10)
1. the document data entry method based on OCR identification and task fragmentation, its step comprises:
1) read the image data of document, and according to the sample information data of document, image data is divided into all kinds of documents;
2) adopt OCR technology to identify the image data of document, obtain the content of each field;
3) according to the setting that is related to of the field contents of OCR identification and each interfield, judge whether each field needs typing;
4) according to OCR coordinate setting to needing the field of typing to carry out cutting, cut into some fragments and be distributed into a plurality of tasks, by internet, carry out fragment type typing.
2. the method for claim 1, is characterized in that: step 2) described employing OCR technology identifies document, comprises image processing, field coordinate setting and field value identification.
3. the method for claim 1, is characterized in that: step 3) according to the OCR identification content of name and two fields of passport NO., whether be that sky judges whether each field needs typing.
4. the method for claim 1, is characterized in that: step 3) according to the OCR identification content of single field, judge whether to need typing: while being identified as sky, field is judged to be sky, does not need typing; While identifying meaningful and content intact, directly get identification content as field value, also do not need to carry out typing; When identification content is imperfect or None-identified goes out content, generate manual entry task.
5. the method for claim 1, it is characterized in that: step 4) by typing field by the difficulty of section content should, the logic configuration of significance level, system and the typing number of times that proof strength dynamically generates needs execution, by once or several, complete typing task.
6. method as claimed in claim 5, is characterized in that: name, sex, birthday, address and phone the field directly once typing task that completes generate; The passport NO. field typing task that makes two bites at a cherry generates, if regressand value checking for the first time correctly, does not need to carry out secondary tasks generation; Organization field is not carried out primary task generation, waits the field typing value that generates for the first time task to return the rear task distribution of directly carrying out organization field when secondary generates.
7. the method as described in any one in claim 1 to 6, it is characterized in that, step 4) also comprise afterwards verification of data and field value integration step, wherein: verification of data is that two record results in same typing field are when inconsistent, the task by artificial judgment input result validity generating, two record results are when all invalid, and the personnel of checking can revise or fill in the right value of field voluntarily; It is by the typing value of field and the value of checking that field value is integrated, and is integrated together the final typing value that generates whole part of each field of document.
8. method as claimed in claim 7, is characterized in that: after field value is integrated, also comprise logical check step, the final typing value of each field, according to the logical check rule separately configuring, is carried out to logic verify and conversion, generate field and become performance number.
9. method as claimed in claim 8, is characterized in that: described logical check is divided into the logical check of individual character section logical check and interfield.
10. method as claimed in claim 8, it is characterized in that, after logical check, carry out finished product inspection and output step, wherein: finished product inspection is that the field that logical check authentication failed is extracted generates artificial finished product inspection task, by professional, check, judge and revise, thereby obtain final document field, becoming performance number; Output is to extract document field final finished value, according to customer demand conversion, outputs in the finished product file of corresponding format, and the finished product that uploads to client's appointment by network service receives catalogue.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410307381.8A CN104077682A (en) | 2014-06-30 | 2014-06-30 | Document data entry method based on OCR and task fragmentization |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410307381.8A CN104077682A (en) | 2014-06-30 | 2014-06-30 | Document data entry method based on OCR and task fragmentization |
Publications (1)
Publication Number | Publication Date |
---|---|
CN104077682A true CN104077682A (en) | 2014-10-01 |
Family
ID=51598927
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410307381.8A Pending CN104077682A (en) | 2014-06-30 | 2014-06-30 | Document data entry method based on OCR and task fragmentization |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104077682A (en) |
Cited By (22)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105005742A (en) * | 2015-07-30 | 2015-10-28 | 四川长虹电器股份有限公司 | Data processing method and data processing system |
CN105022829A (en) * | 2015-07-30 | 2015-11-04 | 四川长虹电器股份有限公司 | System and method for processing data |
CN105243583A (en) * | 2015-09-28 | 2016-01-13 | 四川长虹电器股份有限公司 | Data processing method and data processing system |
CN105550370A (en) * | 2016-01-26 | 2016-05-04 | 平安科技(深圳)有限公司 | Input method and input system |
CN105608452A (en) * | 2014-11-11 | 2016-05-25 | 金蝶软件(中国)有限公司 | Document input method and system |
CN106446901A (en) * | 2016-10-31 | 2017-02-22 | 中国银行股份有限公司 | Method, device and system for entering bank bill |
CN106933870A (en) * | 2015-12-29 | 2017-07-07 | 平安科技(深圳)有限公司 | The record list method and system of data of insuring |
CN107295357A (en) * | 2016-04-01 | 2017-10-24 | 深圳平安综合金融服务有限公司 | Image file data input method, Cloud Server and terminal |
CN108053313A (en) * | 2018-01-02 | 2018-05-18 | 中国工商银行股份有限公司 | Cross-border data processing method of opening an account, apparatus and system |
CN108228320A (en) * | 2016-12-14 | 2018-06-29 | 平安科技(深圳)有限公司 | The method and apparatus of task distribution |
CN108228618A (en) * | 2016-12-14 | 2018-06-29 | 平安科技(深圳)有限公司 | The method and apparatus of document verification of data |
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 |
CN108597565A (en) * | 2018-04-11 | 2018-09-28 | 浙江大学 | It is a kind of that method of calibration is cooperateed with the clinical queuing data of name entity extraction technology based on OCR |
CN108962346A (en) * | 2017-05-22 | 2018-12-07 | 深圳大森智能科技有限公司 | Medical inspection reports electronic method, storage equipment and mobile terminal |
CN109408807A (en) * | 2018-09-11 | 2019-03-01 | 厦门商集网络科技有限责任公司 | The automated testing method and test equipment of OCR recognition correct rate |
CN109800312A (en) * | 2018-12-11 | 2019-05-24 | 广东电网有限责任公司 | A kind of power marketing information management data input system and method |
CN110348804A (en) * | 2019-06-19 | 2019-10-18 | 深圳壹账通智能科技有限公司 | Order entry method, apparatus, computer equipment and storage medium based on AI |
CN110427739A (en) * | 2019-08-09 | 2019-11-08 | 泰康保险集团股份有限公司 | Information Authentication method and device, electronic equipment and computer readable 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) |
CN112215159A (en) * | 2020-10-13 | 2021-01-12 | 苏州工业园区报关有限公司 | International trade document splitting system based on OCR and artificial intelligence technology |
CN112766701A (en) * | 2021-01-13 | 2021-05-07 | 中国外运股份有限公司 | Bill paper and photocopy management method, device, equipment and medium |
CN114896964A (en) * | 2022-05-24 | 2022-08-12 | 中国银行股份有限公司 | Data validity verification method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012029121A (en) * | 2010-07-26 | 2012-02-09 | Seiko Epson Corp | Reading system, image acquisition device, optical reader, method of regulating image acquisition device, and program |
CN102567324A (en) * | 2010-12-14 | 2012-07-11 | 金蝶软件(中国)有限公司 | Receipt field position adjusting method and field position adjuster |
CN102567764A (en) * | 2012-01-13 | 2012-07-11 | 中国工商银行股份有限公司 | Bill certificate and system for improving electronic image recognition efficiency |
CN103246953A (en) * | 2013-04-25 | 2013-08-14 | 天津大学 | Document audit method |
CN103425977A (en) * | 2013-08-05 | 2013-12-04 | 福建亿榕信息技术有限公司 | Financial original certificate image processing method and system |
-
2014
- 2014-06-30 CN CN201410307381.8A patent/CN104077682A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012029121A (en) * | 2010-07-26 | 2012-02-09 | Seiko Epson Corp | Reading system, image acquisition device, optical reader, method of regulating image acquisition device, and program |
CN102567324A (en) * | 2010-12-14 | 2012-07-11 | 金蝶软件(中国)有限公司 | Receipt field position adjusting method and field position adjuster |
CN102567764A (en) * | 2012-01-13 | 2012-07-11 | 中国工商银行股份有限公司 | Bill certificate and system for improving electronic image recognition efficiency |
CN103246953A (en) * | 2013-04-25 | 2013-08-14 | 天津大学 | Document audit method |
CN103425977A (en) * | 2013-08-05 | 2013-12-04 | 福建亿榕信息技术有限公司 | Financial original certificate image processing method and system |
Non-Patent Citations (1)
Title |
---|
蒋春伦: "手写财务报表光电录入系统的设计与实现", 《中国优秀博硕士学位论文全文数据库 (硕士) 信息科技辑》 * |
Cited By (30)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105608452A (en) * | 2014-11-11 | 2016-05-25 | 金蝶软件(中国)有限公司 | Document input method and system |
CN105022829A (en) * | 2015-07-30 | 2015-11-04 | 四川长虹电器股份有限公司 | System and method for processing data |
CN105005742A (en) * | 2015-07-30 | 2015-10-28 | 四川长虹电器股份有限公司 | Data processing method and data processing system |
CN105243583A (en) * | 2015-09-28 | 2016-01-13 | 四川长虹电器股份有限公司 | Data processing method and data processing system |
CN106933870A (en) * | 2015-12-29 | 2017-07-07 | 平安科技(深圳)有限公司 | The record list method and system of data of insuring |
CN105550370A (en) * | 2016-01-26 | 2016-05-04 | 平安科技(深圳)有限公司 | Input method and input system |
CN105550370B (en) * | 2016-01-26 | 2019-03-26 | 平安科技(深圳)有限公司 | Input method and input system |
CN107295357A (en) * | 2016-04-01 | 2017-10-24 | 深圳平安综合金融服务有限公司 | Image file data input method, Cloud Server and terminal |
CN107295357B (en) * | 2016-04-01 | 2021-03-16 | 深圳平安综合金融服务有限公司 | Image file data entry method, cloud server and terminal |
CN106446901A (en) * | 2016-10-31 | 2017-02-22 | 中国银行股份有限公司 | Method, device and system for entering bank bill |
CN108228320A (en) * | 2016-12-14 | 2018-06-29 | 平安科技(深圳)有限公司 | The method and apparatus of task distribution |
CN108228618A (en) * | 2016-12-14 | 2018-06-29 | 平安科技(深圳)有限公司 | The method and apparatus of document verification of data |
CN108228618B (en) * | 2016-12-14 | 2020-07-31 | 平安科技(深圳)有限公司 | Document data checking method and device |
CN108962346B (en) * | 2017-05-22 | 2023-12-15 | 深圳大森智能科技有限公司 | Medical examination report electronization method, storage device and mobile terminal |
CN108962346A (en) * | 2017-05-22 | 2018-12-07 | 深圳大森智能科技有限公司 | Medical inspection reports electronic method, storage equipment and mobile terminal |
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 |
CN108053313A (en) * | 2018-01-02 | 2018-05-18 | 中国工商银行股份有限公司 | Cross-border data processing method of opening an account, apparatus and system |
CN108597565B (en) * | 2018-04-11 | 2021-07-02 | 浙江大学 | Clinical queue data collaborative verification method based on OCR and named entity extraction technology |
CN108597565A (en) * | 2018-04-11 | 2018-09-28 | 浙江大学 | It is a kind of that method of calibration is cooperateed with the clinical queuing data of name entity extraction technology based on OCR |
CN109408807A (en) * | 2018-09-11 | 2019-03-01 | 厦门商集网络科技有限责任公司 | The automated testing method and test equipment of OCR recognition correct rate |
CN109800312A (en) * | 2018-12-11 | 2019-05-24 | 广东电网有限责任公司 | A kind of power marketing information management data input system and method |
CN110348804A (en) * | 2019-06-19 | 2019-10-18 | 深圳壹账通智能科技有限公司 | Order entry method, apparatus, computer equipment and storage medium based on AI |
CN110427739A (en) * | 2019-08-09 | 2019-11-08 | 泰康保险集团股份有限公司 | Information Authentication method and device, electronic equipment and computer readable 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) |
CN112215159A (en) * | 2020-10-13 | 2021-01-12 | 苏州工业园区报关有限公司 | International trade document splitting system based on OCR and artificial intelligence technology |
CN112215159B (en) * | 2020-10-13 | 2021-05-07 | 苏州工业园区报关有限公司 | International trade document splitting system based on OCR and artificial intelligence technology |
CN112766701A (en) * | 2021-01-13 | 2021-05-07 | 中国外运股份有限公司 | Bill paper and photocopy management method, device, equipment and medium |
CN112766701B (en) * | 2021-01-13 | 2024-04-26 | 中国外运股份有限公司 | Method, device, equipment and medium for managing bill of lading paper and photocopying piece |
CN114896964A (en) * | 2022-05-24 | 2022-08-12 | 中国银行股份有限公司 | Data validity verification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104077682A (en) | Document data entry method based on OCR and task fragmentization | |
CN108256074B (en) | Verification processing method and device, electronic equipment and storage medium | |
CN109858453B (en) | Universal multi-engine bill identification system and method | |
CN110348214B (en) | Method and system for detecting malicious codes | |
US11288300B2 (en) | Techniques and components to find new instances of text documents and identify known response templates | |
CN111428599A (en) | Bill identification method, device and equipment | |
CN110990053A (en) | Method for creating and using machine learning scheme template and device | |
CN106339479A (en) | Picture naming method and terminal | |
CN110287098A (en) | Automatically create test script method, server and computer readable storage medium | |
EP2645309A2 (en) | Automatic combination and mapping of text-mining services | |
CN109934255A (en) | A kind of Model Fusion method for delivering object Classification and Identification suitable for beverage bottle recycling machine | |
CN106681854A (en) | Information checking method, device and system | |
CN111752846B (en) | Interface testing method and device | |
CN109885541A (en) | The method and apparatus of information batch processing | |
CN114973290B (en) | Intelligent order examining method and system based on OCR engine | |
CN110458600A (en) | Portrait model training method, device, computer equipment and storage medium | |
CN109409326A (en) | A method of it is kept accounts automatically based on VAT invoice electronic data and generates voucher | |
CN112308727A (en) | Insurance claim settlement service processing method and device | |
CN104064182A (en) | A voice recognition system and method based on classification rules | |
CN114638597A (en) | Intelligent government affair handling application system, method, terminal and medium | |
RU2702967C1 (en) | Method and system for checking an electronic set of documents | |
CN114064893A (en) | Abnormal data auditing method, device, equipment and storage medium | |
CN113704474A (en) | Bank outlet equipment operation guide generation method, device, equipment and storage medium | |
CN106250755A (en) | For generating the method and device of identifying code | |
CN116257404A (en) | Log analysis method and computing device |
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: 20141001 |