Supercharge Your Innovation With Domain-Expert AI Agents!

Image text recognition method for document in natural scene based on deep learning

A text recognition and natural scene technology, applied in neural learning methods, character recognition, character and pattern recognition, etc., can solve the problems of low recognition accuracy, harsh shooting environment requirements, single recognition scene, etc., and achieve the goal of improving recognition accuracy Effect

Active Publication Date: 2022-03-18
XIDIAN UNIV
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to address the above-mentioned deficiencies in the prior art, and propose a natural scene document image character recognition method based on deep learning, which solves the problem that the existing document image character recognition method has strict requirements on the shooting environment, and the recognition scene is single. The problem of low recognition accuracy

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
  • Image text recognition method for document in natural scene based on deep learning
  • Image text recognition method for document in natural scene based on deep learning
  • Image text recognition method for document in natural scene based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] Attached below figure 1 The implementation steps of the present invention are further described.

[0046] Step 1, build image feature extraction module.

[0047] Build a 24-layer feature extraction module, its structure is as follows: first convolutional layer → first pooling layer → second convolutional layer → third convolutional layer → fourth convolutional layer → skip connection layer → fifth volume Product layer→sixth convolutional layer→seventh convolutional layer→skip connection layer→second pooling layer→eighth convolutional layer→ninth convolutional layer→tenth convolutional layer→skip connection layer→eleventh Convolutional layer→twelfth convolutional layer→thirteenth convolutional layer→skip connection layer→third pooling layer→fourteenth convolutional layer→fifteenth convolutional layer→sixteenth convolutional layer→jump connection layer.

[0048] Set the number of convolution kernels in the fourth convolution layer, the seventh convolution layer, the te...

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 discloses a deep learning-based natural scene certificate image character recognition method, the realization steps are: (1) building an image feature extraction module; (2) building a character foreground prediction module; (3) building a character area positioning module; (4) Forming a text positioning network; (5) Constructing a character feature extraction module; (6) Forming a text recognition network; (7) Constructing a text positioning data set; (8) Constructing a text recognition data set; (9) Training a text positioning network ; (10) train the text recognition network; (11) recognize the text in the document image. The present invention overcomes the problem that the existing certificate image character recognition technology has strict requirements on the shooting environment and low recognition accuracy in complex scenes, so that the present invention can accurately recognize the characters in the certificate image in any natural scene.

Description

technical field [0001] The invention belongs to the technical field of image and text processing, and further relates to a text recognition method in natural scene document images based on deep learning in the field of image text recognition technology. The invention can be used to recognize characters in documents (such as ID cards, business licenses, driving licenses, driving licenses) photographed in natural scenes (such as indoor office environments and street scenes). Background technique [0002] Recognizing the text in the image of the certificate is very common and important in many scenarios. For example, in financial scenarios such as remote account opening, online lending, payment verification, etc., we need to identify the name, address, ID number and other information of the user ID card , to check whether the person and certificate are one; the law enforcement of the industrial and commercial department often needs to identify the business name, legal represent...

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): G06V30/413G06V30/146G06V30/148G06V10/82G06V30/10G06N3/04G06N3/08
CPCG06N3/08G06V30/413G06V20/62G06V30/158G06V30/10G06N3/045
Inventor 王晓甜吴嘉诚林亚静石光明齐飞林杰
Owner XIDIAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com