Human Reading Knowledge Inspired Text Line Extraction

Liuan Wang, Seiichi Uchida, Anna Zhu, Jun Sun

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Text in images contains exact semantic information and the text knowledge can be utilized in many image cognition and understanding applications. The human reading habits can provide the clues of text line structure for text line extraction. In this paper, we propose a novel human reading knowledge inspired text line extraction method based on k-shortest paths global optimization. Firstly, the candidate character extraction is reformulated as Maximal Stable Extremal Region (MSER) algorithm on gray, red, blue, and green channels of the target images, and the extracted MSERs are fed into Convolutional Neural Network (CNN) to remove the noise components. Then, the directed graph is built upon the character component nodes with edges inspired by human reading sense. The directed graph can automatically construct the relationship to eliminate the disorder of candidate text components. The text line paths optimization is inspired by the human reading ability in planning of a text line path sequentially. Therefore, the text line extraction problem can be solved using the k-shortest paths optimization algorithm by taking advantage of the human reading sense structure of the directed graph. It can extract the text lines iteratively to avoid the exhaustive searching and obtain global optimized text line number. The proposed method achieves the f-measure of 0.820 and 0.812 on public ICDAR2011 and ICDAR2013 dataset, respectively. The experimental results demonstrate the effectiveness of the proposed human reading knowledge inspired text line extraction method in comparison with state-of-the-art methods This paper presents one human reading knowledge inspired text line extraction method, which approves that the human reading knowledge can benefit the text line extraction and image text discovery.

Original languageEnglish
Pages (from-to)84-93
Number of pages10
JournalCognitive Computation
Volume10
Issue number1
DOIs
Publication statusPublished - Feb 1 2018

Fingerprint

Reading
Directed graphs
Global optimization
Aptitude
Semantics
Cognition
Habits
Noise
Neural networks
Planning

All Science Journal Classification (ASJC) codes

  • Computer Vision and Pattern Recognition
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

Human Reading Knowledge Inspired Text Line Extraction. / Wang, Liuan; Uchida, Seiichi; Zhu, Anna; Sun, Jun.

In: Cognitive Computation, Vol. 10, No. 1, 01.02.2018, p. 84-93.

Research output: Contribution to journalArticle

Wang, Liuan ; Uchida, Seiichi ; Zhu, Anna ; Sun, Jun. / Human Reading Knowledge Inspired Text Line Extraction. In: Cognitive Computation. 2018 ; Vol. 10, No. 1. pp. 84-93.
@article{cfd6b3772b5b400f87e8e2f7483a029a,
title = "Human Reading Knowledge Inspired Text Line Extraction",
abstract = "Text in images contains exact semantic information and the text knowledge can be utilized in many image cognition and understanding applications. The human reading habits can provide the clues of text line structure for text line extraction. In this paper, we propose a novel human reading knowledge inspired text line extraction method based on k-shortest paths global optimization. Firstly, the candidate character extraction is reformulated as Maximal Stable Extremal Region (MSER) algorithm on gray, red, blue, and green channels of the target images, and the extracted MSERs are fed into Convolutional Neural Network (CNN) to remove the noise components. Then, the directed graph is built upon the character component nodes with edges inspired by human reading sense. The directed graph can automatically construct the relationship to eliminate the disorder of candidate text components. The text line paths optimization is inspired by the human reading ability in planning of a text line path sequentially. Therefore, the text line extraction problem can be solved using the k-shortest paths optimization algorithm by taking advantage of the human reading sense structure of the directed graph. It can extract the text lines iteratively to avoid the exhaustive searching and obtain global optimized text line number. The proposed method achieves the f-measure of 0.820 and 0.812 on public ICDAR2011 and ICDAR2013 dataset, respectively. The experimental results demonstrate the effectiveness of the proposed human reading knowledge inspired text line extraction method in comparison with state-of-the-art methods This paper presents one human reading knowledge inspired text line extraction method, which approves that the human reading knowledge can benefit the text line extraction and image text discovery.",
author = "Liuan Wang and Seiichi Uchida and Anna Zhu and Jun Sun",
year = "2018",
month = "2",
day = "1",
doi = "10.1007/s12559-017-9490-4",
language = "English",
volume = "10",
pages = "84--93",
journal = "Cognitive Computation",
issn = "1866-9956",
publisher = "Springer New York",
number = "1",

}

TY - JOUR

T1 - Human Reading Knowledge Inspired Text Line Extraction

AU - Wang, Liuan

AU - Uchida, Seiichi

AU - Zhu, Anna

AU - Sun, Jun

PY - 2018/2/1

Y1 - 2018/2/1

N2 - Text in images contains exact semantic information and the text knowledge can be utilized in many image cognition and understanding applications. The human reading habits can provide the clues of text line structure for text line extraction. In this paper, we propose a novel human reading knowledge inspired text line extraction method based on k-shortest paths global optimization. Firstly, the candidate character extraction is reformulated as Maximal Stable Extremal Region (MSER) algorithm on gray, red, blue, and green channels of the target images, and the extracted MSERs are fed into Convolutional Neural Network (CNN) to remove the noise components. Then, the directed graph is built upon the character component nodes with edges inspired by human reading sense. The directed graph can automatically construct the relationship to eliminate the disorder of candidate text components. The text line paths optimization is inspired by the human reading ability in planning of a text line path sequentially. Therefore, the text line extraction problem can be solved using the k-shortest paths optimization algorithm by taking advantage of the human reading sense structure of the directed graph. It can extract the text lines iteratively to avoid the exhaustive searching and obtain global optimized text line number. The proposed method achieves the f-measure of 0.820 and 0.812 on public ICDAR2011 and ICDAR2013 dataset, respectively. The experimental results demonstrate the effectiveness of the proposed human reading knowledge inspired text line extraction method in comparison with state-of-the-art methods This paper presents one human reading knowledge inspired text line extraction method, which approves that the human reading knowledge can benefit the text line extraction and image text discovery.

AB - Text in images contains exact semantic information and the text knowledge can be utilized in many image cognition and understanding applications. The human reading habits can provide the clues of text line structure for text line extraction. In this paper, we propose a novel human reading knowledge inspired text line extraction method based on k-shortest paths global optimization. Firstly, the candidate character extraction is reformulated as Maximal Stable Extremal Region (MSER) algorithm on gray, red, blue, and green channels of the target images, and the extracted MSERs are fed into Convolutional Neural Network (CNN) to remove the noise components. Then, the directed graph is built upon the character component nodes with edges inspired by human reading sense. The directed graph can automatically construct the relationship to eliminate the disorder of candidate text components. The text line paths optimization is inspired by the human reading ability in planning of a text line path sequentially. Therefore, the text line extraction problem can be solved using the k-shortest paths optimization algorithm by taking advantage of the human reading sense structure of the directed graph. It can extract the text lines iteratively to avoid the exhaustive searching and obtain global optimized text line number. The proposed method achieves the f-measure of 0.820 and 0.812 on public ICDAR2011 and ICDAR2013 dataset, respectively. The experimental results demonstrate the effectiveness of the proposed human reading knowledge inspired text line extraction method in comparison with state-of-the-art methods This paper presents one human reading knowledge inspired text line extraction method, which approves that the human reading knowledge can benefit the text line extraction and image text discovery.

UR - http://www.scopus.com/inward/record.url?scp=85026503745&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85026503745&partnerID=8YFLogxK

U2 - 10.1007/s12559-017-9490-4

DO - 10.1007/s12559-017-9490-4

M3 - Article

AN - SCOPUS:85026503745

VL - 10

SP - 84

EP - 93

JO - Cognitive Computation

JF - Cognitive Computation

SN - 1866-9956

IS - 1

ER -