Single-character text normalization model training method and device, and single-character text recognition method and device

A model training, single-character technology, applied in the field of image recognition, can solve problems such as text pictures that are not suitable for recognizing complex styles, and achieve the effect of accelerating training and convergence, and improving recognition accuracy.

Pending Publication Date: 2020-04-03
上海眼控科技股份有限公司
View PDF14 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Based on this, it is necessary to provide a single-character text normalization model training method, text recognition method and device capable of recognizing complex-style text pictures in view of the above-mentioned problem that the traditional network structure is not suitable for recognizing complex-style text pictures

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
  • Single-character text normalization model training method and device, and single-character text recognition method and device
  • Single-character text normalization model training method and device, and single-character text recognition method and device
  • Single-character text normalization model training method and device, and single-character text recognition method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

[0045] Since traditional OCR tasks usually need to recognize character texts that appear in different fonts, sizes or styles under various complex backgrounds, and the traditional BLSTM+CTC-based network structure is complex, it usually extracts features through convolutional layers to obtain input The feature map of the feature map; then the sequence of the feature map at the width position is regarded as a time series, and the character information is extracted through the BLSTM network; finally, the probability of whether each position is the background or a certain char...

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 relates to a single-character text normalization model training method and device, and a single-character text recognition method and device. The model training method comprises the following steps: acquiring a plurality of single-character sample pictures; normalizing the single-character sample pictures to obtain standard character pictures corresponding to the single-character sample pictures; generating a training data set according to the plurality of single-character sample pictures and standard character pictures in one-to-one correspondence with the plurality of single-character sample pictures; and training a deep learning neural network by using the training data set and a mean square loss function to obtain a single-character text normalization model. The trainingdata set used in training is composed of original data and the standard character pictures which are obtained through normalization processing and have a unified style, so that in the process of training the model, the training and convergence of the model can be accelerated, the model can better learn the essential characteristics of various input texts, and the recognition precision of the modelis further improved.

Description

technical field [0001] The present application relates to the technical field of image recognition, in particular to a single-character text normalization model training method, text recognition method and device. Background technique [0002] With the development of artificial intelligence technology, more and more tedious work is replaced by machines. As an important branch of computer vision, the OCR (Optical Character Recognition) task is widely used in many fields such as bill recognition and automatic text entry. However, usually the original text of this type of task may appear in various complex backgrounds, such as may appear in different styles such as various fonts and sizes. Therefore, it is necessary to use computer vision-related methods to identify the corresponding characters in the text strings located from the pictures for downstream tasks. [0003] As an important branch of artificial intelligence, deep learning has achieved great success in various fiel...

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 Applications(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/04
CPCG06V30/40G06V30/153G06N3/044G06N3/045G06F18/214
Inventor 周康明周枫
Owner 上海眼控科技股份有限公司
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