Scene text extraction in complex images

Hye Ran Byun, Myung Cheol Roh, Kil Cheon Kim, Yeong Woo Choi, Seong Whan Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

5 Citations (Scopus)

Abstract

Text extraction and recognition from still and moving images have many important applications. But, when a source image is an ordinary natural scene, text extraction becomes very complicated and difficult. In this paper, we suggest text extraction methods based on color and gray information. The method using the color image is processed by color reduction, color clustering, and text region extraction and verifications. The method using the gray-level image is processed by edge detection, long line removal, repetitive run-length smearing (RLS), and text region extraction and verifications. Combining two approaches improves the extraction accuracies both in simple and in complex images. Also, estimating skew and perspective of the extracted text regions are considered.

Original languageEnglish
Title of host publicationDocument Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings
EditorsJianying Hu, Ramanujan Kashi, Daniel Lopresti
PublisherSpringer Verlag
Pages329-340
Number of pages12
ISBN (Print)3540440682, 9783540440680
Publication statusPublished - 2002 Jan 1
Event5th International Workshop on Document Analysis Systems, DAS 2002 - Princeton, United States
Duration: 2002 Aug 192002 Aug 21

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2423
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other5th International Workshop on Document Analysis Systems, DAS 2002
CountryUnited States
CityPrinceton
Period02/8/1902/8/21

Fingerprint

Color
Run Length
Edge Detection
Edge detection
Color Image
Skew
Text
Clustering
Line

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Byun, H. R., Roh, M. C., Kim, K. C., Choi, Y. W., & Lee, S. W. (2002). Scene text extraction in complex images. In J. Hu, R. Kashi, & D. Lopresti (Eds.), Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings (pp. 329-340). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2423). Springer Verlag.
Byun, Hye Ran ; Roh, Myung Cheol ; Kim, Kil Cheon ; Choi, Yeong Woo ; Lee, Seong Whan. / Scene text extraction in complex images. Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings. editor / Jianying Hu ; Ramanujan Kashi ; Daniel Lopresti. Springer Verlag, 2002. pp. 329-340 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{9122ef8d5e20462082e9883850f56099,
title = "Scene text extraction in complex images",
abstract = "Text extraction and recognition from still and moving images have many important applications. But, when a source image is an ordinary natural scene, text extraction becomes very complicated and difficult. In this paper, we suggest text extraction methods based on color and gray information. The method using the color image is processed by color reduction, color clustering, and text region extraction and verifications. The method using the gray-level image is processed by edge detection, long line removal, repetitive run-length smearing (RLS), and text region extraction and verifications. Combining two approaches improves the extraction accuracies both in simple and in complex images. Also, estimating skew and perspective of the extracted text regions are considered.",
author = "Byun, {Hye Ran} and Roh, {Myung Cheol} and Kim, {Kil Cheon} and Choi, {Yeong Woo} and Lee, {Seong Whan}",
year = "2002",
month = "1",
day = "1",
language = "English",
isbn = "3540440682",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "329--340",
editor = "Jianying Hu and Ramanujan Kashi and Daniel Lopresti",
booktitle = "Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings",
address = "Germany",

}

Byun, HR, Roh, MC, Kim, KC, Choi, YW & Lee, SW 2002, Scene text extraction in complex images. in J Hu, R Kashi & D Lopresti (eds), Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2423, Springer Verlag, pp. 329-340, 5th International Workshop on Document Analysis Systems, DAS 2002, Princeton, United States, 02/8/19.

Scene text extraction in complex images. / Byun, Hye Ran; Roh, Myung Cheol; Kim, Kil Cheon; Choi, Yeong Woo; Lee, Seong Whan.

Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings. ed. / Jianying Hu; Ramanujan Kashi; Daniel Lopresti. Springer Verlag, 2002. p. 329-340 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2423).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Scene text extraction in complex images

AU - Byun, Hye Ran

AU - Roh, Myung Cheol

AU - Kim, Kil Cheon

AU - Choi, Yeong Woo

AU - Lee, Seong Whan

PY - 2002/1/1

Y1 - 2002/1/1

N2 - Text extraction and recognition from still and moving images have many important applications. But, when a source image is an ordinary natural scene, text extraction becomes very complicated and difficult. In this paper, we suggest text extraction methods based on color and gray information. The method using the color image is processed by color reduction, color clustering, and text region extraction and verifications. The method using the gray-level image is processed by edge detection, long line removal, repetitive run-length smearing (RLS), and text region extraction and verifications. Combining two approaches improves the extraction accuracies both in simple and in complex images. Also, estimating skew and perspective of the extracted text regions are considered.

AB - Text extraction and recognition from still and moving images have many important applications. But, when a source image is an ordinary natural scene, text extraction becomes very complicated and difficult. In this paper, we suggest text extraction methods based on color and gray information. The method using the color image is processed by color reduction, color clustering, and text region extraction and verifications. The method using the gray-level image is processed by edge detection, long line removal, repetitive run-length smearing (RLS), and text region extraction and verifications. Combining two approaches improves the extraction accuracies both in simple and in complex images. Also, estimating skew and perspective of the extracted text regions are considered.

UR - http://www.scopus.com/inward/record.url?scp=84947803663&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947803663&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84947803663

SN - 3540440682

SN - 9783540440680

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 329

EP - 340

BT - Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings

A2 - Hu, Jianying

A2 - Kashi, Ramanujan

A2 - Lopresti, Daniel

PB - Springer Verlag

ER -

Byun HR, Roh MC, Kim KC, Choi YW, Lee SW. Scene text extraction in complex images. In Hu J, Kashi R, Lopresti D, editors, Document Analysis Systems V - 5th International Workshop, DAS 2002, Proceedings. Springer Verlag. 2002. p. 329-340. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).