Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Chinese and Japanese handwritten character recognition method

A technology of handwritten characters and recognition methods, applied in character and pattern recognition, instruments, computer parts, etc.

Inactive Publication Date: 2017-03-22
上海新同惠自动化系统有限公司
View PDF0 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there is no such attempt and research at home and abroad

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
  • Chinese and Japanese handwritten character recognition method
  • Chinese and Japanese handwritten character recognition method
  • Chinese and Japanese handwritten character recognition method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] figure 1 is a system diagram of an apparatus for performing handwriting recognition according to an embodiment. The device can include a handwriting input module, which can be written with an electronic pen or human hand. A display device, such as the screen 102, is configured to receive handwritten input characters, for example, online handwritten characters. The handwritten input characters are input into the handwriting recognition program 104 through the handwriting device, and the recognition program can use one or more modes of offline recognition, online recognition, deep learning recognition or language environment context recognition to perform handwritten characters. identify. The handwriting recognition program 104 may run on the processor 106 , for example, from program instructions stored on the storage medium 108 . The writing recognition program 104 can also use the matching database 110 and the context database 112 for matching recognition. The hand...

Embodiment 2

[0110] Embodiment 2 of another deep learning identification method is:

[0111] see Figure 8 , which mainly includes the following steps:

[0112] Step 800, constructing a multi-layer convolutional neural network, including defining the number of layers constituting the network, the size of the convolution window, and the number of nodes. Generally speaking, the deeper the network, the better the effect, but at the same time the amount of calculation increases. For example, if it contains 2 layers of convolutional layers + 3 layers of fully connected layers, the input is the pixel value of the grayscale image of 28×28 minus the mean value, and each output node of the output layer represents a type of character. For English, there are 62 types ( 26 uppercase letters+26 lowercase letters+10 numbers); the multi-layer convolutional neural network includes multiple convolutional layers and fully connected layers; its input is an image, and its output is a plurality of character ...

Embodiment 3

[0137] Another kind of embodiment 3 is:

[0138] Combining the online recognition candidate set, offline recognition candidate set, language environment context recognition candidate set, and deep learning recognition candidate set, by setting the corresponding weight parameters, the matching results with handwritten input characters are screened out and the results are scored according to Sorting, merging the same candidate characters identified in different candidate sets, performing final sorting, and outputting the final sorted results

[0139] The weight parameter here can be the number of selected candidate characters, for example, four candidate characters can be selected from four different modes to participate in sorting or have different numbers.

[0140] Alternatively, the weight parameter may be multiplied by the scores of candidate characters in different candidate sets according to a certain weight ratio, and finally the corresponding scores are calculated. The ...

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 the field of handwriting character recognition, and particularly relates to a Chinese and Japanese handwritten character recognition method. By using the four recognition methods, such as offline recognition, online recognition, language context recognition and depth learning recognition, handwritten characters are recognized to form a variety of candidate sets. All candidate sets are sorted according to a set weight ratio, and a recognition result is output. The recognition accuracy of Chinese and Japanese handwritten characters is improved. Depth learning and existing text recognition methods are combined together to complement each other and take advantages from each other. An experimental result shows that the recognition accuracy of Chinese and Japanese handwritten characters is greatly improved, and the method has a better beneficial effect compared with the prior art.

Description

technical field [0001] The invention relates to the field of handwritten character recognition, in particular to a recognition method applied to Chinese and Japanese characters. Background technique [0002] With the widespread use of laptops, mobile phones, personal data assistants (PDAs), tablets, smart devices, and virtual reality, more and more people use these smart devices for entertainment, office work, and information acquisition in their daily lives. . In the process of human-computer interaction, it is often necessary to input relevant information, because the input text is carried out through the corresponding input method, for example, Chinese uses the pinyin input method, but due to the large number of Chinese characters, some unfamiliar characters cannot be input through the pinyin input method conduct. Additionally, more portable devices lack full keyboards, where pen-based input is particularly useful in devices that lack full keyboards. [0003] In order ...

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): G06K9/22
CPCG06V30/1423
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 Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
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