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

Segmentation method for improving English online handwritten cursive script recognition rate

A recognition rate and cursive writing technology, applied in character recognition, character and pattern recognition, instruments, etc., can solve problems such as unobvious segmentation points, characters that cannot be clearly separated, and lack of character recognition clues, etc., to narrow down the search path, The effect of improving recognition accuracy

Inactive Publication Date: 2017-04-26
上海新同惠自动化系统有限公司
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Since the separation points between handwritten characters are not obvious, the characters cannot be clearly separated due to the lack of character recognition clues

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
  • Segmentation method for improving English online handwritten cursive script recognition rate
  • Segmentation method for improving English online handwritten cursive script recognition rate
  • Segmentation method for improving English online handwritten cursive script recognition rate

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0062] We evaluate using the publicly available dataset IAM-onDB, which consists of handwritten sentences collected from "smart" whiteboards. It has 4 independent sets: a training set (5364 rows); two validation sets (1438 and 1518 rows); and a test set (3859 rows). We set up the lexicon tree so that it contains only one word (the word string label of the recognized word pattern), and use the word recognizer trained in previous research to recognize each word pattern, and set the recognizable segmentation result as the pattern The character splitting label. Afterwards, we obtained segmentation labels for word patterns in IAM-onDB at the character level.

[0063] We use the training data to train the feature function model. Then, we tested the recognition rate on the test data. Each of these models is trained on 78 symbols. The constructed clue lexicon contains 5562 words (the actual test set lexicon size). The weight parameter Λ is estimated by genetic algorithm validatio...

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 relates to a segmentation method for improving an English online handwritten cursive script recognition rate. The method comprises a step of carrying out preprocessing on words written by a user, and extracting feature points according to certain rule after a series of standard adjustment, a step of running a conditional random field model to extract a neighborhood graph according to the feature points, and evaluating various segmentation paths by using a total energy function, and a step of using a method with the combination of a non-segmentation strategy and a segmentation strategy, in the non-segmentation strategy extension search, using a segmentation candidate lattice constructed in the segmentation strategy to increase a length range to limit a search path, and thus reducing the path range of a synchronous beam search. Compared with the single use of the non-segmentation strategy or the segmentation strategy, the recognition speed is greatly improved on the basis of ensuring the recognition rate.

Description

technical field [0001] The invention relates to a segmentation method for English online handwritten cursive recognition, in particular to a method for automatically segmenting continuously input handwritten English character strings to improve the recognition rate. Background technique [0002] With the large-scale development and maturity of terminal input devices based on stylus or touchpad (such as tablet computers, smart phones, electronic whiteboards and digital pens, etc.), online handwritten character recognition technology is receiving more and more attention. It is crucial to achieve high-quality online handwritten character recognition, especially for text input such as on smartphones, to improve user experience satisfaction. [0003] Since the segmentation points between handwritten characters are not obvious, the characters cannot be clearly separated due to the lack of character recognition clues. A feasible method to overcome this unclear segmentation is call...

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): G06K9/34G06K9/68G06K9/62
CPCG06V30/153G06V30/2445G06V30/248G06V30/10G06F18/24
Inventor 刘建生
Owner 上海新同惠自动化系统有限公司
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