Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image text identification method and device

A text recognition and image technology, applied in the field of image recognition, can solve time-consuming and labor-intensive problems, achieve the effect of solving time-consuming and labor-intensive, avoiding errors, and improving accuracy

Active Publication Date: 2018-05-29
CHINA MOBILEHANGZHOUINFORMATION TECH CO LTD +1
View PDF3 Cites 21 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above technical problems, the embodiments of the present invention expect to provide a method and device for image text recognition, which realizes the effective recognition of single-line text information in natural scenes, and solves the problem of time-consuming and labor-intensive manual segmentation and labeling of massive images. problems, and greatly improved the accuracy of identifying single-line text images, avoiding errors caused by text segmentation in images

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
  • Image text identification method and device
  • Image text identification method and device
  • Image text identification method and device

Examples

Experimental program
Comparison scheme
Effect test

no. 2 example

[0051] In order to better reflect the purpose of the present invention, on the basis of the first embodiment of the present invention, Chinese character recognition is taken as an example for further illustration.

[0052] figure 2 It is a flow chart of the second embodiment of the method for image text recognition of the present invention, the method includes:

[0053] Step 200: Obtain commonly used and newest words from the commonly used corpus to form a word set.

[0054] In actual implementation, commonly used and latest word resources can be obtained from resources such as modern Chinese common word database and popular application program corpus. For example, a total of 202,639 most frequently used and newest words are obtained, which contain 6,699 different Chinese characters.

[0055] Step 201: Simulating the characteristics of natural scenes, expanding each word in the word set into single-line text images with different shapes and backgrounds, and constructing a s...

no. 3 example

[0072] Regarding the method of the embodiment of the present invention, the embodiment of the present invention also provides an image text recognition device. Figure 4 It is a schematic diagram of the composition and structure of the device for image text recognition in the embodiment of the present invention, such as Figure 4 As shown, the device includes: a construction module 400, a processing module 401 and an identification module 402; wherein,

[0073] The construction module 400 is used for constructing a training set of single-line text images.

[0074] The processing module 401 is configured to use the single-line text image training set to train a preset neural network model to obtain a single-line text recognition model.

[0075] The recognition module 402 is configured to use a single-line text recognition model to recognize a single-line text image in a random scene, and obtain recognized text information.

[0076] The construction module 400 is specifically ...

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 an image text identification method comprising the following steps: creating a single-line text image training set; using the single-line text image trainingset to train a preset nerve network model, thus obtaining a single-line text identification model; using the single-line text identification model to identify single-line text images in a random scene, thus obtaining identified text information. Therefore, the method and device can effectively identify single-line text information in natural scenes, thus reducing the manual segmenting costs. The embodiment of the invention also discloses an image text identification device.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to a method and device for image text recognition. Background technique [0002] With the development of science and technology and the progress of society, more and more scientific and technological achievements are being applied to people's daily life and changing people's lives. Among them, the application of image text recognition technology is becoming more and more extensive. However, with the explosive growth of information and the continuous improvement of people's requirements for the accuracy of text recognition in images, the traditional image text recognition technology has been unable to meet the needs of the times. The traditional image text recognition technology mainly has the following problems. [0003] First, the traditional optical character recognition technology (Optical Character Recognition, OCR) is mainly oriented to high-quality document images in image tex...

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): G06K9/00G06K9/62G06N3/08
CPCG06N3/084G06V30/40G06F18/214
Inventor 程耀宋刘一汉杜安安许宝亮
Owner CHINA MOBILEHANGZHOUINFORMATION TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap 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