Unlock instant, AI-driven research and patent intelligence for your innovation.

Character recognition method and device, device for character recognition

A character recognition and character technology, applied in character and pattern recognition, instruments, biological neural network models, etc., can solve the problems of increasing difficulty, segmentation errors, and increasing the error rate corresponding to segmentation errors, and achieves the realization of accuracy and calculation. volume, improve accuracy, and achieve compromise results

Active Publication Date: 2022-04-08
BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD
View PDF16 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, in practical applications, segmentation errors are prone to occur in the process of segmenting line character images into individual character images, which will affect the accuracy of line character recognition results
In particular, for some complex artistic characters or scribbled handwritten characters, there is often a situation where characters are glued together in the line character image, which will increase the difficulty of segmenting the line character image into single-character images, thereby increasing the The error rate corresponding to the segmentation error

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, device for character recognition
  • Character recognition method and device, device for character recognition
  • Character recognition method and device, device for character recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0052]In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0053] An embodiment of the present invention provides a character recognition scheme, which may include: performing size normalization on a line character image containing line characters; using a convolutional neural network to perform feature extraction on the normalized line character image, to obtain the feature information corresponding to the normalized line character image; and perform character recognition on the feature information by using a bidirectional long-short-term memory neural network, so as to obtain a line character recognition result corresponding to the line character image.

[0054] The embodiment of the present invention utilizes the convolutional neural network to perform spatial order-preserving compressio...

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

Embodiments of the present invention provide a character recognition method and device, and a character recognition device, wherein the method specifically includes: normalizing the size of a line character image containing line characters; using a convolutional neural network to normalize Feature extraction is carried out on the line character image after the normalization, to obtain the feature information corresponding to the line character image after the normalization; Utilize the two-way long-short-term memory neural network, carry out character recognition to the feature information, to obtain the line character The line character recognition result corresponding to the image. The embodiment of the present invention can improve the accuracy of the line character recognition result.

Description

technical field [0001] The present invention relates to the technical field of OCR (Optical Character Recognition, Optical Character Recognition), in particular to a character recognition method and device, and a character recognition device. Background technique [0002] With the development of the mobile Internet, the application scenarios of OCR technology have gradually penetrated into the daily life of users from traditional office areas (such as zip codes, book scanning and document faxing). , through OCR technology, the translated Chinese dish names can be displayed on the screen; or, through OCR technology, library books can be converted into e-books, etc. [0003] The current OCR technology can obtain a line of character images from an original image containing characters, and perform character recognition on the line of character images, so as to convert the line of character images into text. The character recognition process for the line of character images may ...

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 Patents(China)
IPC IPC(8): G06V20/62G06V30/146G06V10/26G06V30/148G06V10/46G06V10/82G06N3/04
Inventor 马龙许盛辉徐露露苏雪峰佟子健
Owner BEIJING SOGOU TECHNOLOGY DEVELOPMENT CO LTD