Multi-direction text detection method of natural scene

A text detection and natural scene technology, applied in the field of computer vision, can solve the problems of strong dependence on the training set and poor performance outside the training set, and achieve the effect of suppressing false boundaries and strong adaptability

Active Publication Date: 2015-10-28
SOUTHEAST UNIV
View PDF3 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Rule-based methods are more sensitive to combined text or region breaks, while learning-based methods are highly dependent on the training set and perform poorly outside the training set

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
  • Multi-direction text detection method of natural scene
  • Multi-direction text detection method of natural scene
  • Multi-direction text detection method of natural scene

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] Principle of the present invention is specifically described below in conjunction with accompanying drawing:

[0046] 1. Boundary enhancement MSER region extraction

[0047] The original MSER algorithm uses the region where the area of ​​the extremum region changes to a minimum value as the largest stable extremum region. However, the image boundary is generally blurred, resulting in multiple nested stable extremum regions near the image boundary. The Canny edge detection operator adopts the non-maximum value suppression technology, which effectively suppresses the false boundary, and superimposes the generated boundary into the region. It can be found that the Canny boundary can assist in selecting the best region, thereby eliminating the "false" stable region Such as figure 2 (a) shown.

[0048] Boundary lifting MSER algorithm, recursively pair two regions with father-only child relationship and area change ΔS not exceeding 50% on the stable extremum region compone...

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 provides a multi-direction text detection method of a natural scene. The multi-direction text detection method comprises the following specific steps: 1) extracting a boundary ascension MSER (Maximally Stable Extreme Region), and according to a boundary fit goodness formula, recursively ejecting one area with small boundary fit goodness from two areas which have a father-only child relationship on a stable extreme value area ingredient tree obtained by an original MSER algorithm, wherein the area change [Delta] S of the two areas does not exceed a first threshold value; 2) sorting a character sorting tree area; and 3) carrying out character multi-layer fusion to form text lines, carrying out multi-layer fusion on a sorted character area set which is finally obtained in the step 2), and finally generating the text lines, wherein multiple layers are successively an expansion fusion layer, a free growth layer, a bijection growth layer and a competition layer.

Description

technical field [0001] The invention relates to pattern recognition, image processing, and artificial intelligence related technologies, and belongs to the field of computer vision. Background technique [0002] Text detection in natural scenes is interfered by many factors such as language, scale, font, illumination, contrast, viewing angle, direction, background, incomplete, blurred, broken, etc., and the detection accuracy cannot reach a high level. Text detection in natural scenes has not been well solved until now. The current research is mainly aimed at the detection of English text in the horizontal direction. The detection technology of multi-directional mixed languages ​​is relatively lagging behind. Many detection methods use the character in the horizontal direction as a priori knowledge, so the text detection effect in multiple directions is not ideal (such as [1], [2], [5]), and some detection methods limit the language to English characters, and the trained par...

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): G06K9/00G06K9/62G06K9/66
CPCG06V30/40G06V30/194G06F18/24
Inventor 杨彬夏思宇
Owner SOUTHEAST UNIV
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