Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A method and system for automatically structuring key information of document images

A key information, automatic structure technology, applied in the field of character recognition, can solve the problem of input file type limitation, unable to achieve fully automatic structured output, etc., to reduce interference, improve user experience, and simplify the operation process.

Active Publication Date: 2022-06-21
北京译图智讯科技有限公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] 1. The optical character recognition processing method can only achieve structured output for fixed types of text content, and cannot achieve fully automatic structured output;
[0007] 2. There are restrictions on the input file type, which needs to be a preset file type

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
  • A method and system for automatically structuring key information of document images
  • A method and system for automatically structuring key information of document images
  • A method and system for automatically structuring key information of document images

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0190] According to a specific embodiment of the present invention, with reference to the accompanying drawings, the automatic structuring method for document image key information of the present invention will be described in detail.

[0191] The invention provides an automatic structuring method for document image key information, comprising the following steps:

[0192] S100: Obtain sample image data of the document;

[0193] S300: Perform orientation correction and inclination correction preprocessing on the sample image;

[0194] S400: Use optical character recognition to recognize the text in the sample image, and organize it into text form by line;

[0195] S500: Preprocess the text to obtain text data in units of text blocks;

[0196] S600: Combine the file data in units of text blocks with the model dictionary of the text segmentation model, convert each text block into a number sequence, and obtain the mask sequence, segment sequence and label sequence correspondin...

Embodiment 2

[0201] According to a specific embodiment of the present invention, with reference to the accompanying drawings, the automatic structuring method for document image key information of the present invention will be described in detail.

[0202] The invention provides an automatic structuring method for document image key information, comprising the following steps:

[0203] S100: Obtain sample image data of the document; Step S100 includes the following steps:

[0204] S101: Read file data of files in multiple file formats;

[0205] S102: By setting the ID of each page of file data in the file, the file is divided into single pages, and then each single page is converted into image data.

[0206] S200, load a general text recognition model, a text segmentation model, a text classification model, a text structure extraction model and their configuration files, which are respectively used for text recognition, text segmentation, text classification and text structure extraction;...

Embodiment 3

[0215] According to a specific embodiment of the present invention, with reference to the accompanying drawings, the automatic structuring method for document image key information of the present invention will be described in detail.

[0216] The invention provides an automatic structuring method for document image key information, comprising the following steps:

[0217] S100: Obtain sample image data of the document; Step S100 includes the following steps:

[0218] S101: Read file data of files in multiple file formats;

[0219] S102: By setting the ID of each page of file data in the file, the file is divided into single pages, and then each single page is converted into image data.

[0220] S200, load a general text recognition model, a text segmentation model, a text classification model, a text structure extraction model and their configuration files, which are respectively used for text recognition, text segmentation, text classification and text structure extraction;...

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

No PUM Login to View More

Abstract

The invention provides a method and system for automatically structuring key information of document images, belonging to the technical field of character recognition. The present invention adopts the text in the optical character recognition file, organizes the text into text blocks, then performs text block segmentation through the text segmentation model and the text segmentation model dictionary, and classifies the text blocks through the text classification model and the text classification model dictionary , and finally predict the text block through the prediction model and the prediction model dictionary, and extract the key-value pair data that conforms to the rules according to the prediction results; perform pre-formatted processing on the extracted structured data and then display it. The present invention can realize the identification of any file type, and achieve a structured identification method for automatic structured output results, which is applicable to most common list-type, table-type and other types of voucher reports, and can adapt to the complexity of various voucher reports Scenarios, unified automatic structured output, no need for users to do method configuration and adjustment.

Description

technical field [0001] The invention relates to the technical field of character recognition, in particular to a method and system for automatic structuring of document image key information. Background technique [0002] Computer Character Recognition, commonly known as Optical Character Recognition, English full name Optical Charater Recognition (OCR for short), is a technology that uses optical technology and computer technology to extract the text on the drawing in text form and convert it into a format that humans can understand. In the era of information society, a large amount of bills, forms, and certificate data are generated every day. These data need to be electronically extracted and entered using optical character recognition technology. [0003] With the development of the industry and the maturity of technology, optical character recognition has been applied to many industries, such as sorting and express delivery in the field of logistics, license plate recog...

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
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): G06F16/33G06F16/31G06F16/35G06F40/242G06F40/279G06V30/413G06V30/146G06V30/19G06K9/62G06N3/04G06N3/08
Inventor 王燚王伟饶顶锋陶坚坚刘伟
Owner 北京译图智讯科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products