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

Off-line Chinese character identification method on basis of non-negative matrix factorization

A non-negative matrix decomposition, Chinese character technology, applied in character and pattern recognition, instruments, computer parts and other directions, can solve the problem of reducing the timeliness of character recognition, insufficient use, etc.

Inactive Publication Date: 2011-05-25
广州市伟时信息系统技术有限公司 +1
View PDF0 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, the commonly used font library is GB2312, which includes 6743 Chinese characters, which is not enough for daily use. The GB13000.1-1993 to be promoted includes 21000 Chinese characters, so that the recognition of characters by comparing each character will greatly reduce the The timeliness of character recognition

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
  • Off-line Chinese character identification method on basis of non-negative matrix factorization
  • Off-line Chinese character identification method on basis of non-negative matrix factorization
  • Off-line Chinese character identification method on basis of non-negative matrix factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Since Chinese characters are generally composed of radicals, for example figure 1 As shown, the word "Ling" can be decomposed into two radicals of "Jin" and "Ling". The character recognition can be realized by the comparison of the radicals, that is, it realizes the decomposition of all the character images in a character set image, extracts all the radical images used in the original character set image, and performs radical recognition on the radical image Character comparison, to obtain the radical set of the character set image, and then map each character image composed of radicals to the radical set for comparison, determine which radicals the current character image is composed of, and finally identify the radical The Chinese character corresponding to the character image.

[0037] The present invention provides an off-line Chinese character recognition method based on non-negative matrix decomposition such as figure 2 shown, including the following steps:

...

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 an off-line Chinese character identification method on the basis of non-negative matrix factorization, which comprises the following steps of: carrying out preprocessing on each character image in training set character images and testing set character images which need to be identified so that characters are positioned at the centered positions of the corresponding character image; carrying out non-negative matrix factorization of a character set vector on the preprocessed training set character images to obtain a radical base; carrying out projection on the preprocessed testing set character images on the radical base and obtaining projection coefficients; and identifying corresponding words of the characters in each character image in the testing set character images according to the projection coefficients. The method utilizes shape characteristics and radical characteristics of Chinese characters to realize higher efficiency character identification by a method of carrying out factorization on radicals of the Chinese characters.

Description

technical field [0001] The invention relates to a character image processing method, in particular to an off-line Chinese character recognition method based on non-negative matrix decomposition. Background technique [0002] Character recognition is the theory, method and technology of using computer graphics and image processing technology, combined with the knowledge of probability and statistics, to determine the position of the character image in the character encoding table, and finally determine what the character is, and perform interactive processing. It involves many fields such as computer graphics, image processing, pattern recognition, computer vision, probability statistics, linguistics, etc., and has become a comprehensive technology for studying a series of issues such as data representation, data processing, and decision analysis. Pattern recognition technology can transform scientific data, including data obtained through equipment, images or digital informa...

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): G06K9/62
Inventor 谭军
Owner 广州市伟时信息系统技术有限公司
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

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