CN107516370A - The automatic test and evaluation method of a kind of bank slip recognition - Google Patents
The automatic test and evaluation method of a kind of bank slip recognition Download PDFInfo
- Publication number
- CN107516370A CN107516370A CN201710744296.1A CN201710744296A CN107516370A CN 107516370 A CN107516370 A CN 107516370A CN 201710744296 A CN201710744296 A CN 201710744296A CN 107516370 A CN107516370 A CN 107516370A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- bill
- field
- bank slip
- 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
- Character Discrimination (AREA)
Abstract
The invention discloses a kind of automatic test of bank slip recognition and evaluation method, A, bill comparison template M is made according to business side's demand to the bill in bill test set T;And the field identified according to business side's demand, by required field typing xml document;B, get tickets and be identified according to the bill in test set T into bank slip recognition system successively, and obtain the recognition result of each tested bill, recognition result is write in xml document;C, the field discrimination P of bank slip recognition system is calculatedwWith character identification rate Pc:D, step C result and bill comparison template M are subjected to contrast identification, discrepant field result is exported to text.The present invention only can be being made under the precondition of a template, the test and assessment of bank slip recognition system are realized with computer automation, the time needed for bank slip recognition system application product is substantially reduced, saves out manpower and materials, and there is the advantages that result of calculation speed is fast, and objectivity is high.
Description
Technical field
The present invention relates to picture and text Automatic Measurement Technique field, more particularly to a kind of automatic test of bank slip recognition and comment
Valency method.
Background technology
Various identifying systems (such as identity card identification, fingerprint recognition, the bank slip recognition in picture and text Automatic Measurement Technique field
Deng), as image procossing and area of pattern recognition, an application of computer realm, the crossing domain of artificial intelligence field,
It is a current study hotspot, and actual life requirement.Bank slip recognition as one kind in numerous identifying systems, by
In its demand it is big, have a wide range of application, even more widely studied.
Analyze the forming process of bank slip recognition system application from exploitation to commercialization, it is found that bank slip recognition system is known
The test of other effect consumes a large amount of manpower and materials the iteration that bank slip recognition system is applied is during upgrading, therefore is asked for this
Topic (domestic at present temporarily without the test automation processing to bank slip recognition system identification effect), the invention discloses a kind of automatic
Change the method for tested bill identifying system recognition effect, can effectively reduce manpower and materials, accelerate changing for bill identifying system product
Generation upgrading.
The recognition effect of so-called tested bill identifying system, tested bill identifying system is primarily referred to as in bill
Whether character field identifies that correctly, in general quota invoice has following content, there is invoice title:Shun Feng speed in Sichuan transports limited public affairs
Take charge of Mianyang branch company universal standard invoice, invoice codes:15107158F003, invoice number:Multiple words such as 00004523 ... ..
Section, wherein the content of each specific field is made up of multiple characters, if this field of invoice number is by 00004523 totally 8
Character forms, and the rest may be inferred for remaining field.The recognition effect of tested bill identifying system, exactly see the bill needed for business side
On field (template of a required field is generally provided by business side for identifying system use, this template contains business
Fang Suoxu each bill field) and its character whether identify that correctly accuracy is how high.
Traditional way be it is artificial visually compare each field and whether character correct, not only expend a large amount of manpower and materials,
And subjective, easily error,
The content of the invention
Part in view of the shortcomings of the prior art, it is an object of the invention to provide a kind of automation of bank slip recognition
Test and evaluation method, can effectively reduce manpower and materials, accelerate the iteration upgrading of bank slip recognition system product.
The purpose of the present invention is achieved through the following technical solutions:
The automatic test and evaluation method of a kind of bank slip recognition, its method and step are as follows:
A, bill comparison template M is made according to business side's demand to the bill in bill test set T;And according to business side
The field of demand identification, by required field typing xml document;
B, get tickets and be identified according to the bill in test set T into bank slip recognition system successively, and obtain each tested bill
Recognition result, by recognition result write xml document in;
C, calculated field discrimination and character identification rate:It is assumed that N is shared in single bill comparison template MwIndividual field, i-th
Individual field shares NicIndividual character, the result after bank slip recognition system identification is obtained by character and Field Matching Algorithm, shared
NwrIndividual field identification is correct, and i-th of field shares NicrIndividual character recognition is correct, then can calculate bill by following four formula
The field discrimination P of identifying systemwWith character identification rate Pc:
D, step C result and bill comparison template M are subjected to contrast identification, by discrepant field result export to
Text.
The present invention compared with the prior art, has advantages below and beneficial effect:
The present invention can effectively reduce manpower and materials, accelerate the iteration upgrading of bank slip recognition system product;Energy of the present invention
It is enough that the test and assessment of bank slip recognition system are realized with computer automation in the case where only making the precondition of a template,
The time needed for bank slip recognition system application product is substantially reduced, saves out manpower and materials, and there is result of calculation speed
It hurry up, the advantages that objectivity is high.
Brief description of the drawings
Fig. 1 is the schematic flow sheet of the present invention.
Embodiment
The present invention is described in further detail with reference to embodiment:
Embodiment one
As shown in figure 1, the automatic test and evaluation method of a kind of bank slip recognition, its method and step are as follows:
A, bill comparison template M is made according to business side's demand to the bill in bill test set T;And according to business side
The field of demand identification, by required field typing xml document;
B, get tickets and be identified according to the bill in test set T into bank slip recognition system successively, and obtain each tested bill
Recognition result, by recognition result write xml document in;
C, calculated field discrimination and character identification rate:It is assumed that N is shared in single bill comparison template MwIndividual field, i-th
Individual field shares NicIndividual character, the result after bank slip recognition system identification is obtained by character and Field Matching Algorithm, shared
NwrIndividual field identification is correct, and i-th of field shares NicrIndividual character recognition is correct, then can calculate bill by following four formula
The field discrimination P of identifying systemwWith character identification rate Pc:
D, step C result and bill comparison template M are subjected to contrast identification, by discrepant field result export to
Text.
Embodiment two
The present invention makes to the key technology term occurred and is defined as below:
Bill type:Existing most of bank slip recognition system is identified both for particular kind of bill, such as
Quota invoice is had according to the purpose classification of invoice, network communication machine dismisses ticket etc., and network machine dismisses ticket according to unit of making out an invoice
It is divided into Chinese telecommunications network communication device and dismisses ticket, China Mobile network communication device dismisses ticket, and CHINAUNICOM's network communication device is dismissed
Ticket etc., what is be probably currently known under subdivision has 200 multiclass.
Bill comparison template M:Bill comparison template refers to the part in the bill of the required identification determined by business side
Field and its actual value (being commonly stored in xml document) in bill.Field required for business side is invoice title,
Four invoice codes, invoice number, amount of money fields, bill comparison template include the field required for above-mentioned business side.
Bill field discrimination Pw:Bill field discrimination refers to that bill passes through each word of bank slip recognition system output
The value of section is compared with each field in bill comparison template, and correct field accounts for the ratio of the total field of bill comparison template.
Bill character identification rate Pc:Bill character identification rate refers to that bill passes through all words of bank slip recognition system output
Identify that correct character accounts for the ratio of total character in bill comparison template in section.
Bill test set T:A usual identifying system after commercialization, it is necessary to test its recognition performance, such as
, it is necessary to be tested bank slip recognition performance to assess bank slip recognition system in bank slip recognition system.Bill test set refers to use
In one group of tested bill for examining bank slip recognition system identification performance, usual this group of bill has neither part nor lot in bank slip recognition system
Training process.
As shown in figure 1, Fig. 1 be bank slip recognition system whole bank slip recognition system productization application in status and its
Testing process;The automatic test and evaluation method of a kind of bank slip recognition, its measured step are suddenly as follows:
Step 1, to bill test setTIn bill, according to business side's demand make bill comparison templateM.Manufacturing process
For:The field identified according to business side's demand, by required field typing xml document, when field is more, test set bill
When more, the correctness of typing is examined using the method for cross validation.It is assumed that a total of n times make bill comparison template, note
TheiThe bill comparison template produced isMi, a kind of feasible cross validation method is:
Wherein,The company of expression multiplies symbol, if finding M after i+1 inspectioniBe to and i-th check Mi+1It was found that it is also
To, then cross_validate (Mi,Mi+1)=1, otherwise cross_validate (Mi,Mi+1)=0.
Show that bill comparison template completes as validate_results=1.
Step 2, get tickets successively according to test setTIn bill be identified into bank slip recognition system, and obtain each test
The recognition result of bill, recognition result is write in xml document;
Step 3, calculated field discrimination and character identification rate.It is assumed that N is shared in single bill comparison templatewIndividual field,
I-th of field shares NicIndividual character, the result after bank slip recognition system identification is obtained by character and Field Matching Algorithm, altogether
There is NwrIndividual field identification is correct, theiIndividual field shares NicrIndividual character recognition is correct, then can calculate ticket by following two formulas
According to the field discrimination P of identifying systemwWith character identification rate Pc。
Wherein, | T | represent total bill number in bill test set.
In step 3, character match algorithm calculatesNicrFormula be:
Wherein, find (templateicj, recognitionic) represent i-th of field recognition result
recognitionicThe middle template for searching the field in bill contrast mouldicjThis character whether there is, if in the presence of,
find(templateicj, recognitionic)=1, otherwise find (templateicj, recognitionic)=0.
In step 3, Field Matching Algorithm calculates NwrFormula be:
Wherein, if i-th of field (that is, template of the comparison template of the bill in test setwi) know with bill
I-th of field (that is, recognition of other system identification resultwi) just the same (character matches completely), then compare
(templatewi, recognitionwi)=1, otherwise compare (templatewi, recognitionwi)=0.
Step 4, the variant field of output to text, contrast recognition result and bill comparison template, will be discrepant
Field result is exported to text.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
All any modification, equivalent and improvement made within refreshing and principle etc., should be included in the scope of the protection.
Claims (1)
1. the automatic test and evaluation method of a kind of bank slip recognition, it is characterised in that:Its method and step is as follows:
A, bill comparison template M is made according to business side's demand to the bill in bill test set T;And known according to business side's demand
Other field, by required field typing xml document;
B, get tickets and be identified according to the bill in test set T into bank slip recognition system successively, and obtain the knowledge of each tested bill
Other result, recognition result is write in xml document;
C, calculated field discrimination and character identification rate:It is assumed that N is shared in single bill comparison template MwIndividual field, i-th of field
Shared NicIndividual character, the result after bank slip recognition system identification is obtained by character and Field Matching Algorithm, share NwrIndividual field
Identification is correct, and i-th of field shares NicrIndividual character recognition is correct, then can calculate bank slip recognition system by following four formula
Field discrimination PwWith character identification rate Pc:
<mrow>
<msub>
<mi>N</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
<mi>r</mi>
</mrow>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>N</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
</mrow>
</msub>
</munderover>
<mi>f</mi>
<mi>i</mi>
<mi>n</mi>
<mi>d</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>template</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
<mi>j</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>recognition</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
<mrow>
<msub>
<mi>N</mi>
<mrow>
<mi>w</mi>
<mi>r</mi>
</mrow>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>w</mi>
</msub>
</munderover>
<mi>c</mi>
<mi>o</mi>
<mi>m</mi>
<mi>p</mi>
<mi>a</mi>
<mi>r</mi>
<mi>e</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>template</mi>
<mrow>
<mi>w</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>recognition</mi>
<mrow>
<mi>w</mi>
<mi>i</mi>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>;</mo>
</mrow>
<mrow>
<msub>
<mi>P</mi>
<mi>w</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>T</mi>
<mo>|</mo>
</mrow>
</munderover>
<mfrac>
<msub>
<mi>N</mi>
<mrow>
<mi>w</mi>
<mi>r</mi>
</mrow>
</msub>
<msub>
<mi>N</mi>
<mi>w</mi>
</msub>
</mfrac>
<mo>;</mo>
</mrow>
<mrow>
<msub>
<mi>P</mi>
<mi>c</mi>
</msub>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>j</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mrow>
<mo>|</mo>
<mi>T</mi>
<mo>|</mo>
</mrow>
</munderover>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<msub>
<mi>N</mi>
<mi>w</mi>
</msub>
</munderover>
<mfrac>
<msub>
<mi>N</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
<mi>r</mi>
</mrow>
</msub>
<msub>
<mi>N</mi>
<mrow>
<mi>i</mi>
<mi>c</mi>
</mrow>
</msub>
</mfrac>
<mo>;</mo>
</mrow>
D, step C result and bill comparison template M are subjected to contrast identification, discrepant field result is exported to text text
Part.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710744296.1A CN107516370A (en) | 2017-08-25 | 2017-08-25 | The automatic test and evaluation method of a kind of bank slip recognition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710744296.1A CN107516370A (en) | 2017-08-25 | 2017-08-25 | The automatic test and evaluation method of a kind of bank slip recognition |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107516370A true CN107516370A (en) | 2017-12-26 |
Family
ID=60724284
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710744296.1A Pending CN107516370A (en) | 2017-08-25 | 2017-08-25 | The automatic test and evaluation method of a kind of bank slip recognition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107516370A (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109002768A (en) * | 2018-06-22 | 2018-12-14 | 深源恒际科技有限公司 | Medical bill class text extraction method based on the identification of neural network text detection |
CN109389109A (en) * | 2018-09-11 | 2019-02-26 | 厦门商集网络科技有限责任公司 | The automated testing method and equipment of a kind of this recognition correct rate of OCR full text |
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 |
CN109598837A (en) * | 2018-11-29 | 2019-04-09 | 深圳怡化电脑股份有限公司 | The detection method of financial machine and tool and its distinguishing ability, system and detection service device |
CN111275037A (en) * | 2020-01-09 | 2020-06-12 | 上海知达教育科技有限公司 | Bill identification method and device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4197584A (en) * | 1978-10-23 | 1980-04-08 | The Perkin-Elmer Corporation | Optical inspection system for printing flaw detection |
CN101996438A (en) * | 2010-11-30 | 2011-03-30 | 包钢 | Identifying performance calibration test ticket of fake-detecting currency counting identifier |
CN103440507A (en) * | 2013-09-03 | 2013-12-11 | 北京中电普华信息技术有限公司 | Bill information verifying device and method for verifying bill information |
CN103842991A (en) * | 2011-10-03 | 2014-06-04 | 索尼公司 | Image processing apparatus, image processing method, and program |
CN105574038A (en) * | 2014-10-16 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Text content recognition rate test method and device based on anti-recognition rendering |
-
2017
- 2017-08-25 CN CN201710744296.1A patent/CN107516370A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US4197584A (en) * | 1978-10-23 | 1980-04-08 | The Perkin-Elmer Corporation | Optical inspection system for printing flaw detection |
CN101996438A (en) * | 2010-11-30 | 2011-03-30 | 包钢 | Identifying performance calibration test ticket of fake-detecting currency counting identifier |
CN103842991A (en) * | 2011-10-03 | 2014-06-04 | 索尼公司 | Image processing apparatus, image processing method, and program |
CN103440507A (en) * | 2013-09-03 | 2013-12-11 | 北京中电普华信息技术有限公司 | Bill information verifying device and method for verifying bill information |
CN105574038A (en) * | 2014-10-16 | 2016-05-11 | 阿里巴巴集团控股有限公司 | Text content recognition rate test method and device based on anti-recognition rendering |
Non-Patent Citations (2)
Title |
---|
李翌昕 等: "文本检测算法的发展与挑战", 《信号处理》 * |
虞飞: "机打普通商业***识别***研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109002768A (en) * | 2018-06-22 | 2018-12-14 | 深源恒际科技有限公司 | Medical bill class text extraction method based on the identification of neural network text detection |
CN109389109A (en) * | 2018-09-11 | 2019-02-26 | 厦门商集网络科技有限责任公司 | The automated testing method and equipment of a kind of this recognition correct rate of OCR full text |
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 |
CN109389109B (en) * | 2018-09-11 | 2021-05-28 | 厦门商集网络科技有限责任公司 | Automatic testing method and device for OCR full-text recognition accuracy |
CN109598837A (en) * | 2018-11-29 | 2019-04-09 | 深圳怡化电脑股份有限公司 | The detection method of financial machine and tool and its distinguishing ability, system and detection service device |
CN111275037A (en) * | 2020-01-09 | 2020-06-12 | 上海知达教育科技有限公司 | Bill identification method and device |
CN111275037B (en) * | 2020-01-09 | 2021-06-08 | 上海知达教育科技有限公司 | Bill identification method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107516370A (en) | The automatic test and evaluation method of a kind of bank slip recognition | |
CN105244029B (en) | Voice recognition post-processing method and system | |
CN103336766B (en) | Short text garbage identification and modeling method and device | |
CN111881983B (en) | Data processing method and device based on classification model, electronic equipment and medium | |
Liu et al. | Rethinking attention-model explainability through faithfulness violation test | |
CN109886284B (en) | Fraud detection method and system based on hierarchical clustering | |
CN105389486B (en) | A kind of authentication method based on mouse behavior | |
CN109635105A (en) | A kind of more intension recognizing methods of Chinese text and system | |
CN113297051B (en) | Log analysis processing method and device | |
CN107885849A (en) | A kind of moos index analysis system based on text classification | |
CN109492219A (en) | A kind of swindle website identification method analyzed based on tagsort and emotional semantic | |
CN107491536A (en) | Test question checking method, test question checking device and electronic equipment | |
CN100543735C (en) | File similarity measure method based on file structure | |
Argamon | Computational forensic authorship analysis: Promises and pitfalls | |
CN109255012A (en) | A kind of machine reads the implementation method and device of understanding | |
CN109101483A (en) | A kind of wrong identification method for electric inspection process text | |
CN109976308A (en) | A kind of extracting method of the fault signature based on Laplce's score value and AP cluster | |
CN106446124A (en) | Website classification method based on network relation graph | |
CN106156120A (en) | The method and apparatus that character string is classified | |
Wu et al. | Fine-grained genre classification using structural learning algorithms | |
Sarkar et al. | StRE: Self attentive edit quality prediction in Wikipedia | |
Tutek et al. | Toward practical usage of the attention mechanism as a tool for interpretability | |
Huynh et al. | Towards a benchmark for fact checking with knowledge bases | |
CN116467141A (en) | Log recognition model training, log clustering method, related system and equipment | |
CN105912602A (en) | True-value finding method based on entity attributes |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171226 |
|
RJ01 | Rejection of invention patent application after publication |