A method and device for image text recognition

A text recognition and image technology, applied in character and pattern recognition, instruments, computer parts, etc., can solve the problems of no correction measures, recognition errors, low text recognition accuracy, etc., to meet the recognition needs and improve the accuracy.

Active Publication Date: 2018-09-04
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] Although the above image text recognition method has a strong text recognition ability, but because it is based on the recognition of a single text, it is prone to recognition errors and there is no effective correction measure, and the text recognition accuracy is low

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
  • A method and device for image text recognition
  • A method and device for image text recognition
  • A method and device for image text recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0073] figure 1 The method flowchart of the image text recognition that the embodiment of the present invention provides, such as figure 1 As shown, the method may include the following steps:

[0074] Step 101: Obtain the text area in the image to be recognized.

[0075] The server acquires an image containing text information sent by the mobile terminal, and the image may be an original image captured by the mobile terminal. In this step, the server extracts the text area in the image to be recognized. Or, the image may be that after the mobile terminal captures the original image, extracts the text area in the image to be recognized and then sends the text area in the image to be recognized to the server.

[0076] Existing methods can be used to extract the text area, and the text area can be extracted after removing the image background, but not limited to the following methods:

[0077] Method 1. First, perform color run-length coding according to the color Euclidean d...

Embodiment 2

[0155] Figure 5 The structure diagram of the device for image recognition provided by Embodiment 2 of the present invention, such as Figure 5 As shown, the device includes: an area acquisition unit 500 , a text recognition unit 510 , a position recording unit 520 , a layout analysis unit 530 and a semantic analysis unit 540 .

[0156] First, the area acquisition unit 500 acquires the text area in the image to be recognized, wherein the area acquisition unit 500 can receive the image to be identified sent by the mobile terminal, and extract the text area from the image to be identified; or, the receiving mobile terminal extracts the text area from the image to be identified. and send it to the textarea.

[0157] The character recognition unit 510 respectively recognizes each character block in the character region, and an existing recognition method may be adopted, for example, specifically including: binarizing the character region; dividing the binarized character region i...

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 provides a method and device for recognizing image characters. The method includes the steps of S1, obtaining a character area in an image to be recognized, S2, carrying out recognition on various character blocks in the character area respectively, and recording position information of the character blocks, S3, based on the position information of the character blocks, carrying out layout analysis to obtain statement structure distribution, and S4, based on the statement structure distribution, carrying out correction based on semantic analysis on recognition results of the character blocks to obtain the corrected recognition results. According to the method and device, semantic information among the characters is effectively used for correcting the recognition results of the character blocks, the image character recognition accuracy is improved, and the recognition requirements of users are better met.

Description

【Technical field】 [0001] The invention relates to the field of computer application technology, in particular to a method and device for image and character recognition. 【Background technique】 [0002] With the rapid development of the mobile Internet, the image collected by the camera of the mobile terminal is more and more widely used. Among them, the image character recognition technology recognizes the characters in the image and converts them into text characters, thereby reducing the burden on the user to input the corresponding text information, and making it convenient for the user to store and edit the corresponding text information. However, image text recognition technology is a very complicated technical problem, especially in the case of complex image content, the text recognition accuracy often cannot meet the needs of users. [0003] The existing image text recognition method mainly includes the following steps: [0004] 1) Determine the character area in th...

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): G06K9/20
Inventor 韩钧宇丁二锐吴中勤文林福
Owner BEIJING BAIDU NETCOM SCI & 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