Key information extraction method and system based on semantic segmentation
A key information and semantic segmentation technology, which is applied in neural learning methods, instruments, biological neural network models, etc., can solve problems such as poor timeliness, inaccurate recognition results, and no method for identifying and extracting key information, achieving fast recognition speed , the effect of high accuracy
Pending Publication Date: 2022-01-21
XI AN JIAOTONG UNIV
View PDF5 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The current method of extracting key information of invoices and business licenses mainly uses OCR technology to detect and locate the position of the text first. The method of locating the text is mainly based on the target detection algorithm, which regards the text as the target to be detected and generates a large number of candidate boxes. Then, by classifying, adjusting and screening the candidate frames, the position of the text is finally determined, and then the text is recognized, and finally the keywords are extracted according to the fixed structure of the invoice and the business license, which has the problems of poor timeliness and inaccurate recognition results; in addition, The current existing methods only specifically extract the key information of value-added tax invoices or business licenses, but do not implement a general method of identifying and extracting key information
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0055] In the description of the present invention, it should be understood that the terms "comprising" and "comprising" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude one or more other features, Presence or addition of wholes, steps, operations, elements, components and / or collections thereof.
[0056] It should also be understood that the terminology used in the descriptio...
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 key information extraction method and system based on semantic segmentation. The method comprises the following steps: collecting data to form a learning sample; constructing a deep neural network, and training the neural network by using the data of the sample set; collecting to-be-processed picture data through a terminal or other equipment, and performing enhancement processing on a picture; inputting the processed picture into a pre-trained deep neural network text detector and a text recognizer for text positioning and recognition; the key information is extracted by utilizing information such as fixed formats of invoices and service licenses, position proportions of the key information and the like. Different types of value-added tax invoices and service license key information extraction are integrated in one system, the accuracy and speed of existing recognition are improved, and better systematic service is provided for effectively reducing the workload of manual accounting, reducing the labor cost, achieving finance and tax reimbursement automation, merchant qualification inspection and the like.
Description
technical field [0001] The invention belongs to the technical field of computer digital image processing and pattern recognition, and in particular relates to a method and system for extracting key information based on semantic segmentation. Background technique [0002] Value-added tax invoices and business licenses are often used in daily work. When using them, it is often necessary to check and enter information on the value-added tax invoices and business licenses. The information on the value-added tax invoices and business licenses is more complicated. If Without automatic methods, manual data entry and accounting are required, which is time-consuming and labor-intensive and cannot ensure the accuracy of the entered information. Realize the automatic and rapid identification and entry of general value-added tax invoices and special invoices, which can be applied to application scenarios such as corporate tax accounting and internal reimbursement, which can effectively ...
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
Login to View More IPC IPC(8): G06V10/22G06V10/764G06V10/82G06V30/19G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/044G06N3/045G06F18/24
Inventor赵国帅贾欣悦钱学明
OwnerXI AN JIAOTONG UNIV