CN110414512A - Letter of credit audit terminal - Google Patents
Letter of credit audit terminal Download PDFInfo
- Publication number
- CN110414512A CN110414512A CN201910699442.2A CN201910699442A CN110414512A CN 110414512 A CN110414512 A CN 110414512A CN 201910699442 A CN201910699442 A CN 201910699442A CN 110414512 A CN110414512 A CN 110414512A
- Authority
- CN
- China
- Prior art keywords
- letter
- credit
- clause
- document
- card
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Financial Or Insurance-Related Operations Such As Payment And Settlement (AREA)
Abstract
The embodiment of the invention discloses a kind of letter of credit audit terminals, the letter of credit audit terminal includes: OCR identifying processing unit, for identification and the structural data in letter of credit text image is extracted, wherein the structural data includes: corresponding fund of clause and clause；Intelligence examines card unit, for carrying out soft clause identification to the structural data according to trained soft clause identification model, obtain soft clause recognition result, wherein, the soft clause identification model is to be trained to obtain to preset machine learning model using training sample, and the training sample is the training sample for marking out the corresponding field of each soft clause on the letter of credit structural data for model training and being formed.The present invention solves the identification accuracy occurred when the prior art manually identifies the soft clause in the letter of credit and not can guarantee and the technical issues of low efficiency.
Description
Technical field
The present invention relates to the international business letter of credits to audit technical field, in particular to a kind of letter of credit audit terminal.
Background technique
In international trade, a kind of clearing form of the documentary credit as opposite insurance, using very extensive.China at present
The export trade is mainly also using documentary credit clearing form.Exporter and importer sign sales contract agreement and use credit
Means of account settlement is demonstrate,proved, by importer as applicant, application for credit bank opens to beneficiary (exporter) and is loaded with certain amount of money,
With the file that signs a written pledge for meeting defined document payment within the time limit of certain.After exporter bank receives the letter of credit, need pair
After the letter of credit is audited, careful card result is collectively notified exporter with the original letter of credit.
Meanwhile in practical business there is also exporter and importers before formally opening L/C by letter of credit rough draft text into
Row clause is linked up in advance.Under such business scenario, after exporter receives the letter of credit rough draft of importer offer, exporter is needed
Letter of credit rough draft text is submitted in bank outlets' cabinet face, and letter of credit rough draft is then submitted to bank backstage professional people by site personnel
Member is audited.
It examines card link and mainly audits the letter of credit with the presence or absence of soft clause and document requirement details, prompt exporter to go out
The list processed that mouth quotient and other side carry out after clause modification and delivery guides.Soft Clause in Letter of Credit (Soft Clause), refers to not
Revocable credit, which adds, arranges a kind of clause, prevents exporter from delivering as scheduled, accordingly clause accredited buyer (importer) or
Issuing bank has the one-sided initiative for releasing payment obligations at any time, i.e. buyer fully controls lump-sum transaction, beneficiary (outlet
Quotient) it is in the status of restricted people, and if payment depends entirely on the wish of buyer.Currently, traditional bank still relies on expert to pass through
It tests artificial examine of progress to demonstrate,prove, soft clause identification caused by being easy to appear because of banking personnel specialty technical ability, experience difference is omitted, system
Single guide is indefinite, and manually examines card inefficiency, and the letter of credit is influenced to be notified in time to beneficiary.
At least one of in order to solve the above-mentioned technical problem the present invention, discloses a kind of letter of credit audit terminal.
Summary of the invention
The main purpose of the present invention is to provide a kind of letter of credit audit terminals, to solve the prior art manually to the letter of credit
In the identification accuracy that occurs when being identified of soft clause not can guarantee and the technical issues of low efficiency.
To achieve the goals above, the present invention provides a kind of letter of credit audit terminal, which includes:
OCR identifying processing unit for identification and extracts the structural data in letter of credit text image, wherein described
Structural data includes: corresponding fund of clause and clause；
Intelligence examines card unit, for carrying out soft clause to the structural data according to trained soft clause identification model
Identification, obtains soft clause recognition result, wherein the soft clause identification model is using training sample to preset machine learning
Model, which is trained, to be obtained, the training sample is that each soft item is marked out on the letter of credit structural data for model training
The training sample that the corresponding field of money is formed.
Optionally, the letter of credit audit terminal further include:
Image acquisition units, for acquiring letter of credit text image and sending the letter of credit text image of acquisition to described
OCR identifying processing unit.
Optionally, the letter of credit audit terminal further include:
Document inventory recognition unit, for from extracting document inventory required by the letter of credit in the structural data.
Optionally, the letter of credit audit terminal further include:
Document template acquiring unit, for from the list for obtaining each document in the document inventory in preset document template library
According to template；
Single unit processed, for extracting the corresponding document content-data of the document template from the structural data,
And the document content-data is filled up to formation document sample in the document template.
Optionally, the letter of credit audit terminal further include:
Card result output unit is examined, for according to the soft clause recognition result, the document inventory and the document
In inventory the corresponding document sample of each document generate examine card as a result, and by the careful card result be sent to cloud server into
Row storage.
Optionally, the letter of credit audit terminal further include:
Two dimensional code processing unit, for generating two dimensional code according to the careful card result, so that two dimensional code described in customer scans
The careful card result is downloaded from cloud server.
Optionally, the OCR identifying processing unit is counted when identifying the structural data in the letter of credit text image
Effective text rate that identification generates is calculated, and extracts the letter of credit text when effective text rate is greater than or equal to preset value
Structural data in this image.
Optionally, the OCR identifying processing unit is when effective text rate is less than the preset value by the credit
Card text image is sent to cloud server, so that the cloud server generates artificial examine according to the letter of credit text image
Core task.
Optionally, the letter of credit audit terminal further include:
Remote processing unit, after sending bank for the careful card result and the letter of credit text image
Platform examines card expert server, manually to be veritified to the careful card result.
Optionally, the letter of credit audit terminal further include:
Display unit, for receiving the careful card of the careful card result output unit transmission as a result, and to the careful card
As a result it is shown.
The invention has the benefit that method of the embodiment of the present invention by using machine learning, training generates soft clause
Identification model, and then realize and the soft clause in letter of credit text is identified automatically, recognition efficiency is high.In addition, the present invention is real
Example is applied in training soft clause identification model, by combining manual identified experience to carry out soft clause mark to letter of credit samples of text
Training sample is generated, and carries out model training by a large amount of training samples, keeps trained soft clause identification model identification accurate
Property is higher, solves the technical issues of identification accuracy occurred when the prior art manually carries out soft clause identification not can guarantee.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is the present invention
Some embodiments for those of ordinary skill in the art without creative efforts, can also basis
These attached drawings obtain other attached drawings.In the accompanying drawings:
Fig. 1 is the first structure block diagram of letter of credit audit terminal of the embodiment of the present invention；
Fig. 2 is the second structural block diagram of letter of credit audit terminal of the embodiment of the present invention；
Fig. 3 is the third structural block diagram of letter of credit audit terminal of the embodiment of the present invention；
Fig. 4 is the 4th structural block diagram of letter of credit audit terminal of the embodiment of the present invention；
Fig. 5 is the flow chart that the audit of letter of credit text is carried out using the letter of credit of embodiment of the present invention audit terminal.
Specific embodiment
In order to enable those skilled in the art to better understand the solution of the present invention, below in conjunction in the embodiment of the present invention
Attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is only
The embodiment of a part of the invention, instead of all the embodiments.Based on the embodiments of the present invention, ordinary skill people
The model that the present invention protects all should belong in member's every other embodiment obtained without making creative work
It encloses.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, complete hardware embodiment, complete software embodiment or reality combining software and hardware aspects can be used in the present invention
Apply the form of example.Moreover, it wherein includes the computer of computer usable program code that the present invention, which can be used in one or more,
The computer program implemented in usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) produces
The form of product.
It should be noted that term " includes " and " tool in description and claims of this specification and above-mentioned attached drawing
Have " and their any deformation, it is intended that cover it is non-exclusive include, for example, containing a series of steps or units
Process, method, system, product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include without clear
Other step or units listing to Chu or intrinsic for these process, methods, product or equipment.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the present invention can phase
Mutually combination.The present invention will be described in detail below with reference to the accompanying drawings and embodiments.
It should be noted that used below, term " unit " is the software and/or hardware that predetermined function may be implemented
Combination.Although device described in following embodiment is preferably realized with software, hardware or software and hardware
The realization of combination is also that may and be contemplated.
Fig. 1 is the first structure block diagram of letter of credit audit terminal of the embodiment of the present invention, as shown in Figure 1, the embodiment of the present invention
Letter of credit audit terminal includes: that image acquisition units 1, OCR identifying processing unit 2 and intelligence examine card unit 3.
Image acquisition units 1, for acquiring letter of credit text image and sending institute for the letter of credit text image of acquisition
State OCR identifying processing unit 2.
In embodiments of the present invention, image acquisition units 1 can be camera or scanner, for letter of credit text into
Row is taken pictures or is scanned, and letter of credit text image is obtained.
The letter of credit text can be used for item for exporter and importer before formally opening L/C in embodiments of the present invention
The letter of credit rough draft text that money is linked up, or the letter of credit text that exporter and importer are formally opened.
OCR identifying processing unit 2 for identification and extracts the structural data in letter of credit text image, wherein institute
Stating structural data includes: corresponding fund of clause and clause.
In embodiments of the present invention, OCR identifying processing unit 2 passes through the OCR image recognition method or software of the prior art
It identifies and extracts the structural data in letter of credit text image.
In an embodiment of the present invention, OCR identifying processing unit 2 also can receive user and pass through data line interface, wireless
The letter of credit text image that the modes such as Wifi network transmission are sent, and handled.
In embodiments of the present invention, the core content in letter of credit text is the fund of the letter of credit terms and each clause
(i.e. provision content), the structural data of clause and fund composition letter of credit text.In alternative embodiment of the invention, letter
It may refer to the following table 1 with the letter of credit terms that may include in card text:
Table 1
Intelligence examines card unit 3, for carrying out soft item to the structural data according to trained soft clause identification model
Inscriptions are other, obtain soft clause recognition result, wherein the soft clause identification model is using training sample to preset engineering
Habit model, which is trained, to be obtained, the training sample is each soft to mark out on the letter of credit structural data for model training
The training sample that the corresponding field of clause is formed.
In alternative embodiment of the invention, in order to accurately and efficiently identify the soft clause in letter of credit text, this hair
The bright method using machine learning has trained soft clause identification model.
In alternative embodiment of the invention, the specific trained process of soft clause identification model may refer to following steps 1
To step 5.
Step 1 obtains the letter of credit text for being used for model training.
Step 2 extracts structural data in the letter of credit text.
Step 3, to outpour each soft clause corresponding for corresponding fund subscript of each clause in the structural data extracted
Field forms training sample.
In embodiments of the present invention, soft clause, and then the item in structural data can be summed up according to artificial experience
The corresponding data field of soft clause is marked out on fund.In embodiments of the present invention, the item of some clause is worked as according to artificial experience
When occurring certain contents in fund, it can be determined that the clause is soft clause；Or when respectively there is certain in the fund of multiple clauses
When a little contents, it can determine that these clauses are soft clause.Therefore in mark, the field of these contents is marked out to come, and beat
Upper soft clause label.The soft clause label can number for soft clause, can determine that specially which is soft by soft clause number
Clause.For example, when there is OVERDRAW, LATE SHIPMENT, DC EXPIRED, PCT/ in the fund of numbering of clause 47
When the printed words such as PERCENT, DISCREPANCY, DEDUCT, PENALTY, illustrate that No. 47 clauses are soft clause, the soft clause table
47 X items for showing the letter of credit are that punitive is deducted fees clause, can be marked out the field in mark, and stamp soft clause mark
Label.
In another alternative embodiment of the invention, this step can also be directly by soft clause identification expert in letter of credit text
The enterprising pedestrian's work mark of structural data in this.
Step 4 is trained in preset machine learning model according to the training sample, is generated soft clause and is identified mould
Type.
In alternative embodiment of the invention, can using the FastText model of the prior art, TextCNN model,
ULMFit model, BERT model etc. train the soft clause identification model of the application.In a preferred embodiment of the invention, it uses
FastText model is trained, and the sequence of one word of FastText mode input exports this word sequence and belong to inhomogeneity
Other probability.Word and phrase composition characteristic vector, feature vector in sequence are mapped to middle layer, middle layer by linear transformation
Re-map label.In embodiments of the present invention, a large amount of training samples are input to FastText model and carry out model training.
The identification accuracy of step 5, the soft clause identification model trained according to test sample verifying, so that soft clause is known
The identification accuracy of other model reaches preset value.
In embodiments of the present invention, which can be used as survey to pick out a part from the training sample of generation
Sample sheet, the test sample are used to test the identification accuracy of trained soft clause identification model.When according to training sample training
Out after soft clause identification model, the accuracy of soft clause identification model is verified by test sample, if accuracy deficiency continues to use
Training sample is trained, until the recognition accuracy of the soft clause identification model trained reaches preset value.
In alternative embodiment of the invention, it further includes regular soft clause identification module that intelligence, which examines card unit 3,.In the present invention
In embodiment, certain soft clauses have stronger systematicness, such as MT700 message clause 46A if there is keyword
ISSUED ON APPLICANT'S LETTERHEAD, then explanation is soft clause, which illustrates 46 X items of the letter of credit
Document needs to provide/stamped signature by applicant or its donor.Since these soft clauses have systematicness, it is soft to be referred to as rule
Clause.
In embodiments of the present invention, regular soft clause can be summed up according to artificial experience, generate soft clause template set,
The rule soft clause template includes: clause and corresponding soft clause field contents.Regular soft clause identification module can be according to this
It is soft that soft clause template set carries out rule to the structural data of the letter of credit text extracted by OCR identifying processing unit 2
The identification of item identifies regular soft clause.
In embodiments of the present invention, intelligence examines card unit 3 when carrying out soft clause identification, and it is soft can to first pass through above-mentioned rule
Clause identification module identifies the regular soft clause in outgoing packet, and irregular soft clause is identified using trained soft clause again
Model is identified.In the present embodiment, generate soft clause identification model training sample when only to irregular soft clause into
Rower note, therefore the soft clause identification model generated only identifies irregular soft clause.
As can be seen from the above embodiments, the letter of credit audit terminal of the embodiment of the present invention can upload credit by taking pictures
Text image is demonstrate,proved, OCR identification technology extracts structured message, and the soft clause identification model by loading careful card rule carries out credit
Card audit, automatic export credit demonstrate,prove soft clause information, greatly promote the efficiency and accuracy for examining card.The end of the embodiment of the present invention
End can be supplied to customer manager or customer self-service uses, so that customer manager etc. do not have the personnel that profession examines card qualification
It can complete letter of credit audit or enterprise self-determining completes letter of credit audit, reduce the inconvenience that client faces cabinet.
Fig. 2 is the second structural block diagram of letter of credit audit terminal of the embodiment of the present invention, as shown in Fig. 2, the embodiment of the present invention
Letter of credit audit terminal further include: document inventory recognition unit 4, document template acquiring unit 5 and single unit processed 6.Document is clear
Single recognition unit 4 connect the structural data for obtaining and extracting with the OCR identifying processing unit 2.
Document inventory recognition unit 4, for from extracting document inventory required by the letter of credit in the structural data.
In embodiments of the present invention, include required document clause in letter of credit text, described in the fund of the clause
Document inventory required by the letter of credit.It extracts and identifies by the fund to the required document clause, can determine
Institute's documentary required by the letter of credit.
Document template acquiring unit 5, for from obtaining each document in the document inventory in preset document template library
Document template.
In an embodiment of the present invention, the document mould of different types of letter of credit correlation document is stored in document template library
Plate, the document template comprising documents such as invoice, packing slip, the bill of lading, Certificate of Origin, declaration form, quality inspection proofs.
Single unit 6 processed, for extracting the corresponding document content-data of the document template from the structural data,
And the document content-data is filled up to formation document sample in the document template.
In embodiments of the present invention, single unit 6 processed extracts document according to different document templates from structural data
What template needed fills in data, and then will fill in data filling document template in corresponding domain, generation document sample.After client
It is continuous to be supplemented the production that corresponding document can be rapidly completed in partial information after stock delivery according to the document sample.
Fig. 3 is the third structural block diagram of letter of credit audit terminal of the embodiment of the present invention, as shown in figure 3, the embodiment of the present invention
Letter of credit audit terminal further include: examine card result output unit 7, examine card result output unit 7 respectively with intelligence examine card unit
3, document inventory recognition unit 4 and single unit processed 6 connect.
Card result output unit 7 is examined, for according to the soft clause recognition result, the document inventory and the document
In inventory the corresponding document sample of each document generate examine card as a result, and by the careful card result be sent to cloud server into
Row storage.
In embodiments of the present invention, it is single from the careful card unit 3 of intelligence, the identification of document inventory respectively to examine card result output unit 7
Each document in document inventory needed for member 4 and single unit 6 processed obtain soft clause recognition result, the letter of credit and the document inventory of generation
Corresponding document sample, and summarize generation and examine card result.In embodiments of the present invention, include whether have in soft clause recognition result
Have soft clause, and when there are soft clause each soft clause details.In alternative embodiment of the invention, soft clause
Details include soft clause prompt information, and then subsequent soft clause prompt information be sent to corresponding client (beneficiary goes out
Mouth quotient etc.), to carry out soft clause prompting.In embodiments of the present invention, common soft clause prompt information and its effect can be with
It see the table below 2:
Table 2
Fig. 4 is the 4th structural block diagram of letter of credit audit terminal of the embodiment of the present invention, as shown in figure 4, the embodiment of the present invention
Letter of credit audit terminal further include: two dimensional code processing unit 8, Remote processing unit 9 and display unit 10, at two dimensional code
Reason unit 8, Remote processing unit 9 and display unit 10 are connect with careful card result output unit 7 respectively.
Two dimensional code processing unit 8, for generating two dimensional code according to the careful card result, so that two dimensional code described in customer scans
The careful card result is downloaded from cloud server.
In embodiments of the present invention, the careful card result include: document inventory needed for soft clause details, the letter of credit with
And the corresponding document sample of each document in the document inventory generated.
Remote processing unit 9, for sending bank for the careful card result and the letter of credit text image
Card expert server is examined on backstage, manually to be veritified to the careful card result.
In embodiments of the present invention, Remote processing unit 9 can be responsible for line bank backstage examine card expert carry out into
The careful card opinion of one step seeks advice from confirmation.It selects on-line expert to start line, in real time pushes away document information and intelligent checks result
Send on-line expert；Terminal client video can be carried out with on-line expert or voice is linked up, and answers questions with regard to examining card result progress expert.
Display unit 10, for receiving the careful card of the careful card result output unit transmission as a result, and examining described
Card result is shown.
Fig. 5 is the flow chart that the audit of letter of credit text is carried out using the letter of credit of embodiment of the present invention audit terminal, such as Fig. 5 institute
Show, in one embodiment of this invention, the detailed process of letter of credit text audit is carried out using letter of credit audit terminal of the present invention
It can be with are as follows: obtain the letter of credit image file (letter of credit text image) that user uploads；Using OCR technique to letter of credit image into
Row identification, identifies structured message；Whether the effective text rate for judging that this identification generates is less than preset value；If not less than pre-
If value, then it is careful model of a syndrome (soft clause identification model) and then soft to the letter of credit to start intelligence as inputting according to the data of identification
Clause carries out the detail for the soft clause that automatic identification and registration recognition go out；Clause is required to carry out automatic identification letter of credit document,
Register document inventory list；If being less than preset value, image is uploaded to cloud server, registers manual examination and verification task list,
Card result is examined by manually carrying out remote upload and registering；Careful card result is carried out showing output after obtaining examining card result；It generates
Card result two dimensional code is examined, examines card result convenient for the downloading of scanning input two dimensional code；If user by query, can star careful card result
Long-range card expert's line of examining is seeked advice from.
As can be seen from the above embodiments, the letter of credit audit terminal of the embodiment of the present invention can upload credit by taking pictures
Text image is demonstrate,proved, OCR identification technology extracts structured message, and the soft clause identification model by loading careful card rule carries out credit
Card audit, automatic export credit demonstrate,prove soft clause information, greatly promote the efficiency for examining card, accuracy and auxiliary client (exporter) into
The production of row document.The terminal is supplied to customer manager or customer self-service uses, if user examines card knot to intelligent and portable terminal
Fruit can star online long-range careful card expert and carry out consulting of answering questions, reduce the inconvenience that client faces cabinet there are query.Bank
By providing intelligent terminal of the invention to client, it is voluntarily based on intelligence by client and examines the progress soft clause identification of card terminal and auxiliary
System is single, can play good effect to the synergy inside customer service increased quality and bank.
Obviously, those skilled in the art should be understood that above-mentioned each unit of the invention can be filled with general calculating
It sets to realize, they can be concentrated on a single computing device, or be distributed over a network of multiple computing devices,
Optionally, they can be realized with the program code that computing device can perform, it is thus possible to be stored in storage device
In be performed by computing device, perhaps they are fabricated to each integrated circuit modules or by multiple moulds in them
Block or step are fabricated to single integrated circuit module to realize.In this way, the present invention is not limited to any specific hardware and softwares
In conjunction with.
The foregoing is only a preferred embodiment of the present invention, is not intended to restrict the invention, for the skill of this field
For art personnel, the invention may be variously modified and varied.All within the spirits and principles of the present invention, made any to repair
Change, equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of letter of credit audit terminal characterized by comprising
OCR identifying processing unit for identification and extracts the structural data in letter of credit text image, wherein the structure
Changing data includes: corresponding fund of clause and clause；
Intelligence examines card unit, for carrying out soft clause knowledge to the structural data according to trained soft clause identification model
Not, soft clause recognition result is obtained, wherein the soft clause identification model is using training sample to preset machine learning mould
Type, which is trained, to be obtained, the training sample is that each soft clause is marked out on the letter of credit structural data for model training
The training sample that corresponding field is formed.
2. letter of credit audit terminal according to claim 1, which is characterized in that further include:
Image acquisition units, for acquiring letter of credit text image and sending the OCR for the letter of credit text image of acquisition
Identifying processing unit.
3. letter of credit audit terminal according to claim 1, which is characterized in that further include:
Document inventory recognition unit, for from extracting document inventory required by the letter of credit in the structural data.
4. letter of credit audit terminal according to claim 3, which is characterized in that further include:
Document template acquiring unit, for from the document mould for obtaining each document in the document inventory in preset document template library
Plate；
Single unit processed, for extracting the corresponding document content-data of the document template from the structural data, and will
The document content-data is filled up to formation document sample in the document template.
5. letter of credit audit terminal according to claim 3, which is characterized in that further include:
Card result output unit is examined, for according to the soft clause recognition result, the document inventory and the document inventory
In the corresponding document sample of each document generate and examine card as a result, and sending cloud server for the careful card result and storing up
It deposits.
6. letter of credit audit terminal according to claim 5, which is characterized in that further include:
Two dimensional code processing unit, for according to the careful card result generation two dimensional code, so that two dimensional code described in customer scans is from cloud
Server is held to download the careful card result.
7. letter of credit audit terminal according to claim 1, which is characterized in that the OCR identifying processing unit is identifying
Effective text rate that identification generates is calculated when structural data in the letter of credit text image, and in effective text rate
The structural data in the letter of credit text image is extracted when more than or equal to preset value.
8. letter of credit audit terminal according to claim 7, which is characterized in that the OCR identifying processing unit is described
Cloud server is sent by the letter of credit text image when effective text rate is less than the preset value, so that the cloud takes
Device be engaged according to letter of credit text image generation manual examination and verification task.
9. letter of credit audit terminal according to claim 5, which is characterized in that further include:
Remote processing unit is examined for sending bank backstage for the careful card result and the letter of credit text image
Expert server is demonstrate,proved, manually to be veritified to the careful card result.
10. letter of credit audit terminal according to claim 5, which is characterized in that further include:
Display unit, for receiving the careful card of the careful card result output unit transmission as a result, and to the careful card result
It is shown.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910699442.2A CN110414512A (en) | 2019-07-31 | 2019-07-31 | Letter of credit audit terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910699442.2A CN110414512A (en) | 2019-07-31 | 2019-07-31 | Letter of credit audit terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110414512A true CN110414512A (en) | 2019-11-05 |
Family
ID=68364489
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910699442.2A Pending CN110414512A (en) | 2019-07-31 | 2019-07-31 | Letter of credit audit terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110414512A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046934A (en) * | 2019-12-04 | 2020-04-21 | 中国建设银行股份有限公司 | Method and device for identifying soft clauses of SWIFT message |
| CN111212112A (en) * | 2019-12-17 | 2020-05-29 | 中国建设银行股份有限公司 | Information processing method and device |
| CN111783636A (en) * | 2020-06-30 | 2020-10-16 | 中国工商银行股份有限公司 | OCR-based international reimbursement network application data processing method and device |
| CN111783432A (en) * | 2020-06-30 | 2020-10-16 | 中国工商银行股份有限公司 | Generation method and device of credit certificate examination key point list |
| CN111932369A (en) * | 2020-08-14 | 2020-11-13 | 中国工商银行股份有限公司 | Credit letter opening and closing method, receiving method and device based on block chain |
| CN112801627A (en) * | 2021-02-04 | 2021-05-14 | 台州银行股份有限公司 | Credit document making and auditing method |
| CN113159932A (en) * | 2021-05-14 | 2021-07-23 | 中国工商银行股份有限公司 | Credit certificate verification data processing method and device based on block chain |
| CN114219443A (en) * | 2021-12-16 | 2022-03-22 | 中国建设银行股份有限公司 | Document data processing method, device and equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180129878A1 (en) * | 2013-06-30 | 2018-05-10 | Google Llc | Extracting card data from multiple cards |
| CN108961046A (en) * | 2018-08-09 | 2018-12-07 | 南京怡化信息技术有限公司 | A kind of deposit receipt is opened an account management system and method |
| CN109829692A (en) * | 2019-01-17 | 2019-05-31 | 深圳壹账通智能科技有限公司 | Contract trial method, apparatus, equipment and storage medium based on artificial intelligence |
-
2019
- 2019-07-31 CN CN201910699442.2A patent/CN110414512A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180129878A1 (en) * | 2013-06-30 | 2018-05-10 | Google Llc | Extracting card data from multiple cards |
| CN108961046A (en) * | 2018-08-09 | 2018-12-07 | 南京怡化信息技术有限公司 | A kind of deposit receipt is opened an account management system and method |
| CN109829692A (en) * | 2019-01-17 | 2019-05-31 | 深圳壹账通智能科技有限公司 | Contract trial method, apparatus, equipment and storage medium based on artificial intelligence |
Non-Patent Citations (2)
| Title |
|---|
| 吴争齐: ""用AI为金融行业赋能一览群智发布金融行业智能产品"", 《HTTP://WWW.FCC.COM.CN/ART/KJZX/68/9074.SHTML》 * |
| 汉王: ""汉王OCR金融票据表格单据识别方案"", 《KA.HANWANG.COM.CN/2016/JRBX_0629/43.HTML》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111046934A (en) * | 2019-12-04 | 2020-04-21 | 中国建设银行股份有限公司 | Method and device for identifying soft clauses of SWIFT message |
| CN111046934B (en) * | 2019-12-04 | 2024-04-09 | 中国建设银行股份有限公司 | SWIFT message soft clause recognition method and device |
| CN111212112A (en) * | 2019-12-17 | 2020-05-29 | 中国建设银行股份有限公司 | Information processing method and device |
| CN111783636A (en) * | 2020-06-30 | 2020-10-16 | 中国工商银行股份有限公司 | OCR-based international reimbursement network application data processing method and device |
| CN111783432A (en) * | 2020-06-30 | 2020-10-16 | 中国工商银行股份有限公司 | Generation method and device of credit certificate examination key point list |
| CN111783636B (en) * | 2020-06-30 | 2024-03-29 | 中国工商银行股份有限公司 | OCR-based international balance network application data processing method and device |
| CN111932369A (en) * | 2020-08-14 | 2020-11-13 | 中国工商银行股份有限公司 | Credit letter opening and closing method, receiving method and device based on block chain |
| CN112801627A (en) * | 2021-02-04 | 2021-05-14 | 台州银行股份有限公司 | Credit document making and auditing method |
| CN113159932A (en) * | 2021-05-14 | 2021-07-23 | 中国工商银行股份有限公司 | Credit certificate verification data processing method and device based on block chain |
| CN114219443A (en) * | 2021-12-16 | 2022-03-22 | 中国建设银行股份有限公司 | Document data processing method, device and equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110414512A (en) | Letter of credit audit terminal | |
| CN114862540B (en) | Bill auditing system and method thereof | |
| US20180124437A1 (en) | System and method for video data collection | |
| CN105869032A (en) | Present giving online ordering method and system | |
| CN110781650B (en) | Method and system for automatically generating referee document based on deep learning | |
| CN110443236A (en) | Text will put information extracting method and device after loan | |
| CN106504097A (en) | A kind for the treatment of method and apparatus of assets information | |
| CN108629637A (en) | electronic invoice processing system and electronic invoice processing method | |
| CN110472939A (en) | A kind of supply chain business paper automatic check device, system and method | |
| CN108764942A (en) | A kind of Transaction Information returns control method for anti-counterfeit and system | |
| CN107169856A (en) | A kind of common reserve fund business account checking method and system | |
| CN110399932A (en) | Soft Clause in Letter of Credit recognition methods and device | |
| CN107578328A (en) | A kind of self-aided terminal and method for issuing refund request slip | |
| CN111488458B (en) | Automatic recognition processing method and system for international trade commodity codes | |
| CN111783432A (en) | Generation method and device of credit certificate examination key point list | |
| CN107784585A (en) | Bid business paper processing method, device, computer equipment and storage medium | |
| CN114626102A (en) | Block chain-based electronic certificate transfer method, device, equipment and storage medium | |
| CN106779488A (en) | The proofreading method and system of digital version e-book | |
| CN114445095A (en) | Material detection method, material detection device, storage medium and electronic equipment | |
| Andersen | EDI and data networking in the public sector | |
| CN110503382A (en) | Retail medicine company shops application system, method and device applied to mobile terminal | |
| CN114913026A (en) | Insurance front-end system, equipment and medium | |
| CN114092085A (en) | Electronic gift generation, distribution and execution method and platform | |
| CN113313531A (en) | E-commerce trial evaluation and recommendation system based on user requirements | |
| CN109558449A (en) | Data processing platform (DPP) and data processing method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |
|
| RJ01 | Rejection of invention patent application after publication |