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

Text-independent end-to-end handwriting recognition method based on deep learning

A text-independent, deep learning technology, applied in neural learning methods, character and pattern recognition, instruments, etc., can solve the problems of the writer's recognition accuracy needs to be improved, no handwriting samples are found, and the recognition accuracy is not high. The effect of fitting, generalization ability enhancement, and recognition rate improvement

Active Publication Date: 2016-08-24
CHONGQING AOXIONG INFORMATION TECH
View PDF4 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, due to the lack of data and no particularly good features of handwritten samples, the accuracy of writer recognition needs to be improved
The previous recognition methods are based on pattern recognition and other related computer image processing technologies. Some feature expressions and distance calculations are artificially determined, and character features are artificially extracted. The accuracy of recognition is not high.

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
  • Text-independent end-to-end handwriting recognition method based on deep learning
  • Text-independent end-to-end handwriting recognition method based on deep learning
  • Text-independent end-to-end handwriting recognition method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0044] The present invention mainly solves the identification of online text writers and its specific realization, adopts the preprocessing method of segmenting online text and randomly removing stroke segments, and establishes a complete text-independent end-to-end handwriting based on deep learning recognition methods. The present invention has no restrictions on the type of characters input by the user, nor on the text, and can allow the user to write free text to the greatest extent. The overall process is as follows: figure 1 shown.

[0045] see figure 1, the present invention includes the following four processes: A, the preprocessing process of online handwritten text; B, the deep neural network model training process of known writer samples; C, the calculation path integral feature; identification process. Specifically, it is first necessary to resample the text lines of online handwritten long texts to become online text lines with equal sampling point spacing, and...

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 text-independent end-to-end handwriting recognition method based on deep learning, comprising the following steps: A, preprocessing an online handwritten text to generate a pseudo character sample; B, calculating the path integral feature image of the pseudo character sample; C, training a deep neural network model of a sample of a known writer; and D, using the deep neural network model in step C to automatically recognize a sample of an unknown writer. Through the method, online text lines can be processed automatically, there is no need to extract character features manually, and text-independent online writer recognition is realized efficiently.

Description

technical field [0001] The invention belongs to the field of deep learning and artificial intelligence, and in particular relates to a feature learning technique for distinguishing writers from text-independent online handwritten documents handwritten by computer users. Background technique [0002] Handwriting is an important basis for identity verification and legal evidence collection. In recent years, the popularization of touch mobile terminals and writing tablets has increased the attention of online handwriting identification. However, due to the lack of data and no particularly good features of handwritten samples, the accuracy of writer identification needs to be improved. The previous recognition methods are based on pattern recognition and other related computer image processing technologies. Some feature expressions and distance calculations are artificially determined, and character features are artificially extracted. The accuracy of recognition is not high. ...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06N3/08
CPCG06N3/08G06V30/347G06F18/214
Inventor 金连文杨维信刘曼飞
Owner CHONGQING AOXIONG INFORMATION TECH
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

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com