Relaxational extracting method for character recognition in scene images.

Jun Ohya, Akio Shio, Shigeru Akamatsu

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

1 Citation (Scopus)

Abstract

An effective extraction algorithm for character recognition in scene images is developed. Character candidate regions are detected by an image-segmentation method based on adaptive thresholding and by evaluating gray-level difference between adjacent regions. Character pattern candidates are obtained by associating the detected regions according to their positions and gray levels and evaluating the aspect ratio of the rectangle that circumscribes the associated regions. The character-recognition process selects character pattern candidates with high similarities to any categories in a dictionary (high similarity patterns). Highly promising experimental results have been obtained using the method on 100 images acquired under uncontrolled lighting. This algorithm can be applied to read characters of any size and format under various lighting conditions.

Original languageEnglish
Title of host publicationProc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit
Place of PublicationSilver Spring, MD, United States
PublisherPubl by IEEE
Pages424-429
Number of pages6
ISBN (Print)0818608625
Publication statusPublished - 1988
Externally publishedYes

Fingerprint

Character recognition
Lighting
Glossaries
Image segmentation
Aspect ratio

ASJC Scopus subject areas

  • Engineering(all)

Cite this

Ohya, J., Shio, A., & Akamatsu, S. (1988). Relaxational extracting method for character recognition in scene images. In Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit (pp. 424-429). Silver Spring, MD, United States: Publ by IEEE.

Relaxational extracting method for character recognition in scene images. / Ohya, Jun; Shio, Akio; Akamatsu, Shigeru.

Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit. Silver Spring, MD, United States : Publ by IEEE, 1988. p. 424-429.

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

Ohya, J, Shio, A & Akamatsu, S 1988, Relaxational extracting method for character recognition in scene images. in Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit. Publ by IEEE, Silver Spring, MD, United States, pp. 424-429.
Ohya J, Shio A, Akamatsu S. Relaxational extracting method for character recognition in scene images. In Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit. Silver Spring, MD, United States: Publ by IEEE. 1988. p. 424-429
Ohya, Jun ; Shio, Akio ; Akamatsu, Shigeru. / Relaxational extracting method for character recognition in scene images. Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit. Silver Spring, MD, United States : Publ by IEEE, 1988. pp. 424-429
@inproceedings{92345fee403b4928a5426148952fcdb8,
title = "Relaxational extracting method for character recognition in scene images.",
abstract = "An effective extraction algorithm for character recognition in scene images is developed. Character candidate regions are detected by an image-segmentation method based on adaptive thresholding and by evaluating gray-level difference between adjacent regions. Character pattern candidates are obtained by associating the detected regions according to their positions and gray levels and evaluating the aspect ratio of the rectangle that circumscribes the associated regions. The character-recognition process selects character pattern candidates with high similarities to any categories in a dictionary (high similarity patterns). Highly promising experimental results have been obtained using the method on 100 images acquired under uncontrolled lighting. This algorithm can be applied to read characters of any size and format under various lighting conditions.",
author = "Jun Ohya and Akio Shio and Shigeru Akamatsu",
year = "1988",
language = "English",
isbn = "0818608625",
pages = "424--429",
booktitle = "Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit",
publisher = "Publ by IEEE",

}

TY - GEN

T1 - Relaxational extracting method for character recognition in scene images.

AU - Ohya, Jun

AU - Shio, Akio

AU - Akamatsu, Shigeru

PY - 1988

Y1 - 1988

N2 - An effective extraction algorithm for character recognition in scene images is developed. Character candidate regions are detected by an image-segmentation method based on adaptive thresholding and by evaluating gray-level difference between adjacent regions. Character pattern candidates are obtained by associating the detected regions according to their positions and gray levels and evaluating the aspect ratio of the rectangle that circumscribes the associated regions. The character-recognition process selects character pattern candidates with high similarities to any categories in a dictionary (high similarity patterns). Highly promising experimental results have been obtained using the method on 100 images acquired under uncontrolled lighting. This algorithm can be applied to read characters of any size and format under various lighting conditions.

AB - An effective extraction algorithm for character recognition in scene images is developed. Character candidate regions are detected by an image-segmentation method based on adaptive thresholding and by evaluating gray-level difference between adjacent regions. Character pattern candidates are obtained by associating the detected regions according to their positions and gray levels and evaluating the aspect ratio of the rectangle that circumscribes the associated regions. The character-recognition process selects character pattern candidates with high similarities to any categories in a dictionary (high similarity patterns). Highly promising experimental results have been obtained using the method on 100 images acquired under uncontrolled lighting. This algorithm can be applied to read characters of any size and format under various lighting conditions.

UR - http://www.scopus.com/inward/record.url?scp=0024131764&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024131764&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0818608625

SP - 424

EP - 429

BT - Proc CVPR 88 Comput Soc Conf on Comput Vision and Pattern Recognit

PB - Publ by IEEE

CY - Silver Spring, MD, United States

ER -