Detecting text in natural scenes with stroke width transform

Boris Epshtein; Eyal Ofek; Yonatan Wexler

doi:10.1109/CVPR.2010.5540041

Detecting text in natural scenes with stroke width transform

Boris Epshtein, Eyal Ofek, Yonatan Wexler

Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We present a novel image operator that seeks to find the value of stroke width for each image pixel, and demonstrate its use on the task of text detection in natural images. The suggested operator is local and data dependent, which makes it fast and robust enough to eliminate the need for multi-scale computation or scanning windows. Extensive testing shows that the suggested scheme outperforms the latest published algorithms. Its simplicity allows the algorithm to detect texts in many fonts and languages.

Original language	English
Title of host publication	2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Publisher	IEEE
Pages	2963-2970
Number of pages	8
ISBN (Print)	978-1-4244-6983-3
DOIs	https://doi.org/10.1109/CVPR.2010.5540041
Publication status	Published - 18 Jun 2010
Event	2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - San Francisco, CA, USA Duration: 13 Jun 2010 → 18 Jun 2010

Conference

Conference	2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Period	13/06/10 → 18/06/10

Keywords

Layout
Optical character recognition software
Pixel
Computer vision
Filter bank
Engines
Image segmentation
Colored noise
Geometry
Robustness

Access to Document

10.1109/CVPR.2010.5540041

https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=5540041

Cite this

@inproceedings{2509a3517887432594bfd8be6da24204,

title = "Detecting text in natural scenes with stroke width transform",

abstract = "We present a novel image operator that seeks to find the value of stroke width for each image pixel, and demonstrate its use on the task of text detection in natural images. The suggested operator is local and data dependent, which makes it fast and robust enough to eliminate the need for multi-scale computation or scanning windows. Extensive testing shows that the suggested scheme outperforms the latest published algorithms. Its simplicity allows the algorithm to detect texts in many fonts and languages.",

keywords = "Layout, Optical character recognition software, Pixel, Computer vision, Filter bank, Engines, Image segmentation, Colored noise, Geometry, Robustness",

author = "Boris Epshtein and Eyal Ofek and Yonatan Wexler",

year = "2010",

month = jun,

day = "18",

doi = "10.1109/CVPR.2010.5540041",

language = "English",

isbn = "978-1-4244-6983-3",

pages = "2963--2970",

booktitle = "2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition",

publisher = "IEEE",

note = "2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition ; Conference date: 13-06-2010 Through 18-06-2010",

}

TY - GEN

T1 - Detecting text in natural scenes with stroke width transform

AU - Epshtein, Boris

AU - Ofek, Eyal

AU - Wexler, Yonatan

PY - 2010/6/18

Y1 - 2010/6/18

N2 - We present a novel image operator that seeks to find the value of stroke width for each image pixel, and demonstrate its use on the task of text detection in natural images. The suggested operator is local and data dependent, which makes it fast and robust enough to eliminate the need for multi-scale computation or scanning windows. Extensive testing shows that the suggested scheme outperforms the latest published algorithms. Its simplicity allows the algorithm to detect texts in many fonts and languages.

AB - We present a novel image operator that seeks to find the value of stroke width for each image pixel, and demonstrate its use on the task of text detection in natural images. The suggested operator is local and data dependent, which makes it fast and robust enough to eliminate the need for multi-scale computation or scanning windows. Extensive testing shows that the suggested scheme outperforms the latest published algorithms. Its simplicity allows the algorithm to detect texts in many fonts and languages.

KW - Layout

KW - Optical character recognition software

KW - Pixel

KW - Computer vision

KW - Filter bank

KW - Engines

KW - Image segmentation

KW - Colored noise

KW - Geometry

KW - Robustness

UR - https://ieeexplore.ieee.org/document/5540041/

U2 - 10.1109/CVPR.2010.5540041

DO - 10.1109/CVPR.2010.5540041

M3 - Conference contribution

SN - 978-1-4244-6983-3

SP - 2963

EP - 2970

BT - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

PB - IEEE

T2 - 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition

Y2 - 13 June 2010 through 18 June 2010

ER -