Handwritten character recognition method and device, storage medium and terminal

A text recognition and text technology, applied in the field of text recognition, can solve problems such as low recognition accuracy, achieve the effect of improving robustness, reducing workload, and enriching sample types and styles

Inactive Publication Date: 2021-02-19
上海芯翌智能科技有限公司
View PDF16 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, at present, offline handwritten text recognition mainly refers to the re

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
  • Handwritten character recognition method and device, storage medium and terminal
  • Handwritten character recognition method and device, storage medium and terminal
  • Handwritten character recognition method and device, storage medium and terminal

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0038] As mentioned above, in the current prior art, the accuracy rate for handwritten character recognition is relatively low. Only research has found that in the process of character recognition, decoding and recognition is performed word by word, and the errors in the previous character recognition will be propagated and accumulated to the latter, thereby affecting the overall recognition accuracy and resulting in a lower accuracy rate of character recognition.

[0039] In order to solve the above problems, in the embodiment of the present invention, when performing character recognition on handwritten characters, text detection is performed on images of handwritten characters to be recognized to obtain row text images corresponding to each row of text. Perform feature extraction on the row text image to obtain the first feature map, perform attention area positioning on each row text image, and obtain the second feature map with the probability of occurrence of each text, a...

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 a handwritten character recognition method and device, a storage medium and a terminal. The method comprises the steps of obtaining a handwritten character image to be recognized; performing text detection on the handwritten character image to be recognized to obtain a row text image corresponding to each row text; inputting each row of text images into a character recognition model for character recognition to obtain a character recognition result corresponding to the handwritten character image to be recognized, and performing the following processing on the row of text images by the character recognition model: performing feature extraction on each row of text images to obtain a corresponding first feature map; carrying out attention region positioning on each row of text images to obtain a second feature map carrying the occurrence position probability of each character; fusing the first feature map and the second feature map to obtain a fused feature map; and performing encoding and decoding based on the fused feature map to obtain a character recognition result corresponding to each row of text images, thereby obtaining a character recognition result.According to the scheme, the handwritten character recognition accuracy can be improved.

Description

technical field [0001] Embodiments of the present invention relate to the field of character recognition, and in particular, to a handwritten character recognition method and device, a storage medium, and a terminal. Background technique [0002] As an important technology in the field of computer vision, text recognition can help humans automatically enter text information in reality. Text recognition is specifically divided into scene text recognition and handwritten text recognition. Scene text recognition is mainly for printed text recognition, and handwritten text recognition is mainly for handwritten text recognition. [0003] Handwritten character recognition can be divided into online handwritten character recognition and offline handwritten character recognition. Online handwritten text recognition usually uses a touchpad as an input device. According to the character position and inclination of the input Chinese character, even features such as stroke order can be...

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): G06K9/00G06K9/46G06K9/62
CPCG06V40/33G06V10/40G06F18/253G06F18/214
Inventor 姬东飞陈新泽黄冠都大龙
Owner 上海芯翌智能科技有限公司
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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap