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

Character recognition method and device

A text recognition and text technology, applied in the field of text recognition, can solve the problems of low recognition accuracy, high error rate of text segmentation or merging, and affect the accuracy of recognition results, etc., to achieve low text merging error rate and improve recognition accuracy Effect

Inactive Publication Date: 2017-09-15
BEIJING SANKUAI ONLINE TECH CO LTD
View PDF5 Cites 13 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, in the current recognition method, there is a high error rate in the segmentation or merging of text, which will affect the accuracy of the final recognition result
That is, the current text recognition method has 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
  • Character recognition method and device
  • Character recognition method and device
  • Character recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0062] A character recognition method provided by an embodiment of the present invention is introduced in detail.

[0063] refer to figure 1 , shows a flowchart of steps of a character recognition method in an embodiment of the present invention.

[0064] In step 101, the text image is segmented using multiple segmentation methods to obtain multiple image segmentation results, and the multiple segmentation regions included in each image segmentation result respectively include at least one object to be recognized.

[0065] The above-mentioned text image may be an image containing text information, and the contained text information may be text in multiple languages ​​such as Chinese, English, and French.

[0066] The above-mentioned segmentation method may include various ways of segmenting the text in the text image, for example, using the gap between each text in a whole string of text as the segmentation point to divide it into multiple texts; Text gaps are the multiple t...

Embodiment 2

[0082] Another character recognition method provided by the embodiment of the present invention is introduced in detail.

[0083] refer to figure 2 , shows a flow chart of steps of another character recognition method in an embodiment of the present invention.

[0084] In step 201, the character image is segmented using multiple segmentation methods to obtain multiple image segmentation results, and the multiple segmentation regions included in each image segmentation result respectively include at least one object to be recognized.

[0085] Optionally, the step 201 may include the following sub-steps:

[0086] Sub-step S11, marking multiple candidate segmentation points on the text image.

[0087] In sub-step S12, a plurality of candidate segmentation point sets are respectively formed according to different selected candidate segmentation points.

[0088] In sub-step S13 , according to each set of candidate segmentation points, the character image is segmented respective...

Embodiment 3

[0155] A character recognition device provided by an embodiment of the present invention is introduced in detail.

[0156] refer to image 3 , shows a structural block diagram of a character recognition device in Embodiment 3 of the present invention.

[0157] The device may include:

[0158] The image segmentation result acquisition module 301 is used to segment the text image using multiple segmentation methods to obtain multiple image segmentation results, and the multiple segmentation regions included in each image segmentation result include at least one object to be identified respectively;

[0159] The comprehensive weight value statistics module 302 is used to count the comprehensive weight value of each image segmentation result according to the object to be recognized in the segmentation area, wherein the comprehensive weight value includes each segmentation area in the image segmentation result Statistical values ​​of text feature matching degree and shape feature...

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 character recognition method and device. The method comprises the steps that multiple segmentation modes are utilized to segment a character image to obtain multiple image segmentation results, wherein each of multiple segmented regions included in each image segmentation result comprises at least one to-be-recognized object; statistical analysis is performed on a synthetic weight value of all the image segmentation results according to the to-be-recognized objects in the segmented regions, wherein the synthetic weight value comprises statistical values of character feature matching degrees and shape feature matching degrees of all the segmented regions in the image segmentation results; an optimal image segmentation result is screened according to the synthetic weight value of all the image segmentation results; and target characters corresponding to all the segmented regions in the optimal image segmentation result are adopted to serve as recognition results. According to the embodiment, the recognition accuracy of the character image is improved.

Description

technical field [0001] The present invention relates to the field of character recognition, in particular to a character recognition method, a character recognition device, a computer device, and a computer-readable storage medium. Background technique [0002] In the field of text recognition, especially for the recognition of text images containing a whole string of text, it is usually necessary to divide the whole string of text into multiple individual texts, and recognize each text separately. [0003] When segmenting, the image is usually segmented according to the segmentation path composed of multiple segmentation points, and the segmented area may contain text formed by merging multiple characters. The text is recognized, and the recognition result can be obtained. [0004] However, in the current recognition method, there is a high error rate in text segmentation or merging, which will affect the accuracy of the final recognition result. That is, the current char...

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): G06K9/34G06K9/62
CPCG06V30/153G06V30/287G06F18/2411
Inventor 魏晓明
Owner BEIJING SANKUAI ONLINE TECH CO LTD
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