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

Image character recognition system and method

A text recognition and image technology, applied in the field of image processing, can solve the problems of low accuracy of text recognition model, difficulty in labeling new training data, and high cost of manual labeling, so as to reduce the workload of manual labeling, reduce repeated labeling, and improve the model. The effect of precision

Pending Publication Date: 2019-06-25
上海兑观信息科技技术有限公司
View PDF9 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0009] Embodiments of the present invention provide a system and method for image text recognition, adopting an intelligent hybrid labeling method, which solves the problem that the current intelligent text understanding only uses part of the label samples in the text detection model and text recognition model with low accuracy, and the new training data is marked Difficult, high cost of manual labeling

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 character recognition system and method
  • Image character recognition system and method
  • Image character recognition system and method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] According to one or more embodiments, such as figure 1 As shown, an image text recognition method is used to identify the text information contained in the image, including the following steps:

[0025] S101, use the existing or marked image text data set to train the basic text detection model and text recognition model; the text detection model here can be applied to various image text detection models, including but not limited to: for standard rectangles Frame-marked models (such as Connectionist Text Proposal Network, etc.), for character detection models with a certain rotation angle (such as RRPN, EAST, DMP-Net, etc.), for character detection models of arbitrary shapes (such as Total-Text, TextSnake, etc.) . The text recognition model can be applied to various image text recognition models, including but not limited to: Attention-based deep network models, CTC-based deep network models, etc. If a preliminary model is trained using the existing image datasets th...

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

An image character recognition system comprises a detection model and / or a recognition model. The detection model is used for detecting the position of an area where characters in the image are located. The recognition model is used for extracting character information in the image. An operation module calls the detection module and / or the recognition module to detect and recognize the charactersin the image and obtain a preliminary labeling result. A sample selection module is used for selecting part or all of the samples from the preliminary labeling result. A labeling correction module isused for correcting the labeling result selected by the sample selection module to obtain a refined labeling result, and the refined labeling result is used for continuously training and optimizing the detection model and / or the recognition model.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to an image character recognition system and method. Background technique [0002] Image text intelligent understanding technology refers to the use of artificial intelligence methods and models to process and analyze images, and determine the specific location and content of text in the image. The recognition and judgment of the specific position and content of the text in the image is helpful to understand the entire image, and at the same time provides a basis for structured extraction of image content and judgment of key information to solve practical problems. For example, it helps realize sensitive text information identification, intelligent OCR document identification, etc. [0003] Image text intelligent understanding technology can include 2 steps, including: [0004] 1. Extract the area where the text content in the image is located, this process is also called...

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): G06K9/32
Inventor 周钊郑莹斌叶浩
Owner 上海兑观信息科技技术有限公司
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