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

Online handwritten Uighur word data enhancement method

A word and data technology, applied in the field of online handwritten Uyghur word data enhancement, can solve problems such as unsatisfactory results and changes in the type of samples, so as to improve the overall performance, enhance the enhancement effect, and solve the effect of data shortage

Active Publication Date: 2018-10-16
XINJIANG UNIVERSITY
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

that might be a downside of it
Be careful when using the track point discarding method on text that is sensitive to delayed strokes, because this method may discard those delayed strokes that consist of only one point, resulting in a change in the type of the sample
Some methods lead to suboptimal results if directly applied to the overall handwritten word trajectory

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
  • Online handwritten Uighur word data enhancement method
  • Online handwritten Uighur word data enhancement method
  • Online handwritten Uighur word data enhancement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0060] The technical solution of the present invention will be described in further detail below in conjunction with the accompanying drawings and embodiments.

[0061] 1. Uyghur Online Handwritten Data Enhancement Method Based on the Combination of Multiple Algorithms

[0062] Such as figure 1 shown in . According to the advantages and disadvantages of different data enhancement methods, the data enhancement algorithms proposed or adopted by the present invention are respectively implemented on individual strokes and overall samples.

[0063] 1.1 The length of the stroke trajectory changes randomly

[0064] Variations in flat segments in handwritten word trajectories are most likely to vary sample trajectory length, as well as width and height of the overall sample shape. In Uyghur handwritten words, the variation of horizontal flat segmentation has more influence on the overall sample than the variation of vertical flat segmentation. Therefore, the present invention only...

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 online handwritten Uighur word data enhancement method. The method analyzes the writing characteristics of handwritten Uighur words, proposes an online handwritten data enhancement algorithm with handwritten trajectories randomly variable lengths, combines with a variety of data enhancement algorithms suitable for online handwritten words, and achieves online handwrittenUighur word data enhancement. The data enhancement method combined with a variety of algorithms is obviously effective, and a small number of original samples can be used for constructing more effective forged samples of different handwriting styles while readability is ensured. The data enhancement method has strong versatility and can be used as a direct reference for other handwritten data enhancement researches.

Description

technical field [0001] The invention belongs to the technical field of handwriting recognition, and relates to an online handwritten Uyghur word data enhancement method, in particular to an online handwritten Uyghur word data enhancement method based on the combination of multiple algorithms. Background technique [0002] Handwriting recognition is a hot topic in the field of pattern recognition and machine learning. With the advancement of machine learning research, using machine learning algorithms to construct and train handwriting recognition models has become a common method in the field of handwriting recognition. In machine learning research, the larger the amount of training data, the stronger the generalization ability of the trained model. This situation is even more evident in deep learning research. The amount of data is an important factor that directly affects the generalization ability of deep models. The size of the data volume is directly related to the r...

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
IPC IPC(8): G06K9/68
CPCG06V30/2455G06V30/293
Inventor 吾加合买提·司马义玛依热·依布拉音艾斯卡尔·艾木都拉
Owner XINJIANG UNIVERSITY
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