Unlock instant, AI-driven research and patent intelligence for your innovation.
Model correction method and correction system
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A model correction and model technology, applied in character and pattern recognition, instrument, character recognition, etc., can solve problems such as lack of error correction mechanism, and achieve the effect of improving recognition accuracy and recognition rate
Inactive Publication Date: 2019-10-15
SUNING CLOUD COMPUTING CO LTD
View PDF7 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
If it is necessary to improve the recognition success rate, a large number of samples are required for marking training, and there is a lack of an error correction mechanism based on the entire pipeline of the business system to improve the overall recognition rate
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more
Image
Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
example 1
[0053] "ocrName": "totleMoney",
[0054] "ocrValue": "198.00",
[0055] "businessFlag": "1",
[0056] "ocrReusltDet": false,
[0057] "businessValue": "300.00"
[0058] In Example 1, the amount is wrongly recognized, the business data is 300, and the model recognition is 198. Therefore, the marker model identifies errors.
example 2
[0060] "ocrName": "ticketsMoney",
[0061] "ocrValue": "192.23",
[0062] "businessFlag": "1",
[0063] "ocrReusltDet": true,
[0064] "businessValue": "192.23"
[0065] In example 2, identify the amount, the business data is 192.23, and the model identification is 192.23. Therefore, the marker model is identified correctly.
[0066] After the field information is generated from the data information, the system can easily find the model data information marked as wrong from the field information. For example, in the above two examples, the field information of "ocrReusltDet": false is searched directly. When the model data information marked as error is found, the corresponding error data information is obtained. Finally, according to the model data information marked as error, the general character recognition result with coordinates is restored. Such as image 3 As shown, the wrong model data information on the right side of the figure is restored to the general text...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
PUM
Login to View More
Abstract
The invention discloses a model correction method and correction system, which can improve the recognition rate of a model. The model correction method comprises the steps that data information is collected from a picture, and the data information comprises correct data information that service data and model identification data are consistent and error data information that the service data and the model identification data are inconsistent; judging whether the error data information is a model identification error or not, and if so, entering the next step; and performing model antisense according to the error data information, automatically marking, and training a model.
Description
technical field [0001] The invention belongs to the field of model correction, and in particular relates to a model correction method and a correction system. Background technique [0002] At present, most of the invoice reimbursement, scanning, and review in the financial shared office field still adopt manual review. Financial personnel are faced with the process of receiving, sorting, scanning, reviewing, and replying to physical invoices every day. The entire process line is boring, and there are a lot of repetitive manual operations, especially the bill review. It is necessary to repeatedly compare whether the reimbursement process information reported by employees is consistent with the bill information, which is time-consuming and laborious. [0003] The traditional ocr recognition is based on the attention_ocr technology to complete the general recognition model. Through the continuous marking training of the marking platform, the text recognition results from left ...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More
Application Information
Patent Timeline
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.