Handwritten text line character segmentation method and identification method
A text line and character technology, applied in the field of handwritten text line character recognition and handwritten text line character segmentation, can solve the problems of indistinct character boundaries, time-consuming, and unsatisfactory segmentation effects, and achieve convenient character recognition processing. Effect
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0063] The present invention provides a fast unconstrained handwritten text line segmentation method, which can accurately segment the text line. Three common situations in naturally written Chinese text lines: naturally separated characters, overlapping characters, and glued characters are processed separately. For naturally separated characters, use the histogram projection method to segment them; for overlapping characters, use the method of setting isolation points in the overlapping area to segment them; for glued characters, first detect the glued points, and then Split the characters at the glue point, and finally divide them as overlapping characters. The method of the present invention can ensure high segmentation accuracy while realizing rapid segmentation of text lines.
[0064] The present invention proposes a character segmentation technical method for three common situations in non-unconstrained natural writing Chinese text lines: naturally separated characters, ov...
Embodiment 2
[0101] This embodiment discloses a method for recognizing characters in a handwritten text line. The method includes the following steps:
[0102] Step S10: input a text line;
[0103] Step S20: Divide the naturally separated characters;
[0104] Step S30: Estimate the width of each character based on the result obtained in step S20, and determine whether the character width is greater than the preset threshold T; if yes, go to step S40; otherwise, go to step S70;
[0105] Step S40: Perform overlapping character segmentation;
[0106] Step S50: Estimate the width of each character based on the result obtained in step S40, and determine whether the character width is greater than the preset threshold T; if yes, go to step S60; otherwise, go to step S70;
[0107] Step S60: Perform segmentation of glued characters;
[0108] Step S70: output the segmentation result;
[0109] Step S80: Recognize the characters obtained from the output segmentation results respectively, and output the recognitio...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
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