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

A method and system for text detection in natural scenes based on self-heuristic strategy

A text detection and natural scene technology, applied in the field of image recognition, can solve the problems of high complexity and low detection rate of text detection, and achieve the effect of improving detection efficiency and reducing search area

Active Publication Date: 2019-06-25
HUNAN NORMAL UNIVERSITY
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The present invention provides a natural scene text detection method and system based on a self-heuristic strategy, and its purpose is to overcome the problems of low detection rate and high complexity of natural scene text detection in the prior art

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 system for text detection in natural scenes based on self-heuristic strategy
  • A method and system for text detection in natural scenes based on self-heuristic strategy
  • A method and system for text detection in natural scenes based on self-heuristic strategy

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0136] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0137] Such as figure 1 As shown, a natural scene text detection method based on a self-heuristic strategy includes the following steps:

[0138] Step 1: Convert the input color natural scene image to grayscale natural scene image;

[0139] In the embodiment of the present invention, the resolution of the input color natural scene image is 1280×960, and the formula used in the embodiment of the present invention to convert the color image to the grayscale natural scene image is:

[0140] I gray (i,j)=0.2989×I R (i,j)+0.5870×I G (i,j)+0.1140×I B (i,j)

[0141] Among them, I R (i, j) represents the red component value of the pixel whose coordinate position is (i, j) in the color natural scene image, I G (i, j) represents the green component value of the pixel at the coordinate position (i, j) in the color natural scene image, I B (i, j) represent...

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 natural scene text detection method and system based on a self-heuristic strategy. A two-layer cascade filtering mechanism is established to judge the text area and the background area. Considering that the text area has dark background and light text in the grayscale natural scene image Or two modes of light background and dark characters, the contrast type of the text area is judged so as to unify the contrast of the text area in the grayscale natural scene image into the mode of dark background and light text; As the seed text, a self-heuristic search strategy is constructed for each seed text in its adjacent area, and the trained deep neural network is used to detect missing text in the form of a sliding detection window to improve the recall rate of text detection. Finally, the adjacent text is Words are separated into units. The natural scene text detection method and system disclosed by the present invention have high practical value for realizing automatic understanding of natural scene text semantics.

Description

technical field [0001] The invention belongs to the field of image recognition, in particular to a method and system for detecting characters in natural scenes based on a self-heuristic strategy. Background technique [0002] As an extremely important information carrier, text plays a great role in promoting the process of human civilization. As a visual carrier of human language, text widely exists in all aspects of human society. In recent years, with the rapid development of computer vision technology, humans have managed to enable computers to simulate human visual understanding mechanisms, analyze captured images and understand image content independently, so as to obtain important clues in images; natural scenes include All kinds of scene content, as an important scene content, text has abstract description ability and high generalization ability. Detecting and recognizing text in images will help to understand image content and obtain important information clues. In...

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): G06K9/00G06K9/32G06K9/34G06K9/62G06K9/72
CPCG06V30/40G06V20/62G06V30/158G06V10/267G06V30/274G06F18/22G06F18/24G06F18/25
Inventor 王润民钱盛友杨剑锋邹孝
Owner HUNAN NORMAL UNIVERSITY
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