Text recognition method, electronic equipment and computer readable medium

A text recognition and text image technology, applied in the computer field, can solve the problems of inability to adapt to the change of handwritten text style, poor handwritten text recognition effect, etc.

Pending Publication Date: 2021-04-09
BEIJING CENTURY TAL EDUCATION TECH CO LTD
View PDF6 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the existing text recognition models that have a good recognition effect on printed texts have

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
  • Text recognition method, electronic equipment and computer readable medium
  • Text recognition method, electronic equipment and computer readable medium
  • Text recognition method, electronic equipment and computer readable medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] refer to figure 1 , shows a flowchart of steps of a text recognition method according to Embodiment 1 of the present application.

[0036] The text recognition method of the present embodiment comprises the following steps:

[0037] Step 101, perform feature extraction on a text image to be recognized to obtain corresponding image features.

[0038] The text recognition method in the embodiment of the present application is applicable to the recognition of various texts, for example, it can be used to recognize text images containing only printed text; it can also be used to recognize text images containing only handwritten text; It can also be used to recognize text images that contain both printed text and handwritten text; in addition, the text recognition method in the embodiment of the present application can also be used to recognize long texts that contain a large amount of text. It is especially applicable to text recognition of text images containing handwrit...

Embodiment 2

[0052] Embodiment 2 of the present application is based on the solution of Embodiment 1. Optionally, in one embodiment of the present application, performing self-attention calculation processing based on image features in step 102 to obtain corresponding feature encoding vectors may include:

[0053] Perform fully connected feature extraction processing on image features to obtain triplet vectors; perform self-attention calculation processing based on triplet vectors to obtain corresponding feature encoding vectors.

[0054] Since the self-attention calculation process is mainly realized based on the triplet vector, and the result obtained in step 101 is the image feature, therefore, the full connection feature extraction process can be performed on the image feature first (usually, Full connection operation can be performed on the image features) to obtain the triplet vector, so that the subsequent self-attention calculation process can be performed based on the triplet v...

Embodiment 3

[0068] refer to figure 2 , shows a flowchart of steps of a text recognition method according to Embodiment 3 of the present application.

[0069] In this embodiment, the text recognition method is executed based on a preset neural network model.

[0070] see image 3 , image 3 A schematic structural diagram of the neural network model provided by the embodiment of the present application, the neural network model may include: an image feature extraction part; a self-attention part and a position encoding part connected in parallel after the image feature extraction part; and a self-attention part and The stitching part connected with the position encoding part; the semantic feature extraction part connected with the stitching part.

[0071] in:

[0072] The image feature extraction part is used to perform feature extraction on the text image to be recognized, and output corresponding image features. Optionally, the image feature extraction part can be realized by CNN. ...

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 embodiment of the invention discloses a text recognition method, electronic equipment and a computer readable medium. The text recognition method comprises the steps: performing feature extraction on a text image to be recognized, and obtaining corresponding image features; performing self-attention calculation processing based on the image features to obtain corresponding feature coding vectors; carrying out character position enhancement processing based on the image features to obtain corresponding character position coding vectors; splicing the feature coding vectors and the character position coding vectors, and performing semantic feature extraction on the spliced coding vector to obtain a semantic feature vector; and decoding the semantic feature vector to obtain a corresponding text character. According to the scheme of the embodiment of the invention, the character recognition accuracy and the character bit sequence of the finally decoded character are more accurate.

Description

technical field [0001] The embodiments of the present application relate to the field of computer technology, and in particular, to a text recognition method, electronic equipment, and a computer-readable medium. Background technique [0002] Text recognition is a technology that detects images containing text and obtains its corresponding text information. [0003] When the current text recognition technology recognizes an image containing text, it will be interfered by external factors, such as image clarity, image exposure, different text fonts, etc., resulting in inaccurate recognition results. Taking text fonts as an example, in many scenarios, such as in students' homework or test papers, there will be handwritten text in addition to printed text. However, because handwritten texts do not have the standardization of printed texts, the styles of handwritten texts of different students are quite different. Therefore, the existing text recognition models that have a goo...

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): G06K9/34G06K9/46G06N3/04G06N3/08
CPCG06N3/08G06V30/153G06V10/462G06V30/10G06N3/045
Inventor 姜明刘霄熊泽法
Owner BEIJING CENTURY TAL EDUCATION TECH CO LTD
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