Text real-time positioning recognition method based on deep learning attention mechanism

A real-time positioning and deep learning technology, applied in the field of text recognition, can solve the problems of difficulty in locating text regions, poor robustness of character segmentation, and low recognition accuracy, so as to avoid inaccurate detection, improve recognition accuracy, and speed up training. effect of speed

Inactive Publication Date: 2019-05-14
安徽艾睿思智能科技有限公司
View PDF0 Cites 19 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Therefore, OCR recognition in the prior art has the problems of difficulty in locating text regions in complex backgrounds, poor character

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 real-time positioning recognition method based on deep learning attention mechanism
  • Text real-time positioning recognition method based on deep learning attention mechanism
  • Text real-time positioning recognition method based on deep learning attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0039] The preferred embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so as to define the protection scope of the present invention more clearly.

[0040] see figure 1 , the embodiment of the present invention includes:

[0041] A text real-time location recognition method based on deep learning attention mechanism, comprising the following steps:

[0042] S1: Build a text image acquisition system, collect training samples and manually mark them, and establish an OCR data set;

[0043] For the collected training samples, delete invalid images and manually label them. Randomly select 80,000 images as the test set, and about 20,000 remaining images as the training set. Use text files to store the labeling information of each picture, and use endpoints Named in the form of coordinates plus te...

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 discloses a text real-time positioning recognition method based on a deep learning attention mechanism. The method comprises the following steps: establishing a text image acquisition system, collecting a training sample, carrying out manual marking, establishing an OCR data set, carrying out the preprocessing of an image according to the characteristics of the OCR data set, then providing a deep saliency attention network to locate a text region, and enabling a text to be separated from a complex background; and finally, identifying the text by utilizing deep convolution cyclicattention to realize real-time detection and identification of the text image. According to the method, the characters do not need to be segmented, the detection and recognition precision of the OCR system can be effectively improved, and the method has good real-time performance and mobility and has wide application prospects.

Description

technical field [0001] The invention relates to the technical field of text recognition, in particular to a real-time text positioning and recognition method based on a deep learning attention mechanism. Background technique [0002] OCR technology is the abbreviation of Optical Character Recognition (Optical Character Recognition). It converts the text of various bills, newspapers, books, manuscripts and other printed materials into image information through optical input methods such as scanning, and then uses text recognition technology to convert the image information. Enter technology for computers that can be used. It can be applied to the input and processing fields of bank bills, a large amount of text materials, archives, and copywriting. It is suitable for automatic scanning identification and long-term storage of a large number of bill forms in banking, taxation and other industries. [0003] The prior art generally includes four steps: image preprocessing, text...

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): G06K9/20G06K9/62
CPCG06V10/22G06F18/241
Inventor 吴仕莲曹洋郑志刚
Owner 安徽艾睿思智能科技有限公司
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