Grayscale feature combination in recognition based segmentation for degraded text string recognition

Jun Sun; Yoshinobu Hotta; Katsuhito Fujimoto; Katsuyama Yutaka; Satoshi Naoi

Grayscale feature combination in recognition based segmentation for degraded text string recognition

Jun Sun, Yoshinobu Hotta, Katsuhito Fujimoto, Katsuyama Yutaka, Satoshi Naoi

国際情報通信研究センター

研究成果: Paper

3 被引用数 (Scopus)

抄録

Grayscale feature is very effective for degraded character recognition. While many papers focus on different feature extraction algorithms on single character recognition, few deals with the impact of the selected feature on segmentation. For recognition-based segmentation, a good recognition performance on single character may not always have good performance on segmentation. In this paper, two types of grayscale feature, the R-Feature and the S-Feature, are proposed based on dual-eigenspace decomposition. The RFeature is suitable for single character recognition. The SFeature is suitable for text string segmentation. These two feature are combined to further improve the performance for degraded Japanese text string recognition.

本文言語	English
ページ	39-44
ページ数	6
出版ステータス	Published - 2005
イベント	1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005 - Seoul, Korea, Republic of 継続期間: 2005 8 29 → 2005 8 29

Conference

Conference	1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005
Country	Korea, Republic of
City	Seoul
Period	05/8/29 → 05/8/29

ASJC Scopus subject areas

Computer Vision and Pattern Recognition
Software

文献へのアクセス

フィンガープリント「Grayscale feature combination in recognition based segmentation for degraded text string recognition」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル

@conference{13f9f354ccbc45108d65a3b4fe30ad33,

title = "Grayscale feature combination in recognition based segmentation for degraded text string recognition",

abstract = "Grayscale feature is very effective for degraded character recognition. While many papers focus on different feature extraction algorithms on single character recognition, few deals with the impact of the selected feature on segmentation. For recognition-based segmentation, a good recognition performance on single character may not always have good performance on segmentation. In this paper, two types of grayscale feature, the R-Feature and the S-Feature, are proposed based on dual-eigenspace decomposition. The RFeature is suitable for single character recognition. The SFeature is suitable for text string segmentation. These two feature are combined to further improve the performance for degraded Japanese text string recognition.",

author = "Jun Sun and Yoshinobu Hotta and Katsuhito Fujimoto and Katsuyama Yutaka and Satoshi Naoi",

year = "2005",

language = "English",

pages = "39--44",

note = "1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005 ; Conference date: 29-08-2005 Through 29-08-2005",

}

TY - CONF

T1 - Grayscale feature combination in recognition based segmentation for degraded text string recognition

AU - Sun, Jun

AU - Hotta, Yoshinobu

AU - Fujimoto, Katsuhito

AU - Yutaka, Katsuyama

AU - Naoi, Satoshi

PY - 2005

Y1 - 2005

N2 - Grayscale feature is very effective for degraded character recognition. While many papers focus on different feature extraction algorithms on single character recognition, few deals with the impact of the selected feature on segmentation. For recognition-based segmentation, a good recognition performance on single character may not always have good performance on segmentation. In this paper, two types of grayscale feature, the R-Feature and the S-Feature, are proposed based on dual-eigenspace decomposition. The RFeature is suitable for single character recognition. The SFeature is suitable for text string segmentation. These two feature are combined to further improve the performance for degraded Japanese text string recognition.

AB - Grayscale feature is very effective for degraded character recognition. While many papers focus on different feature extraction algorithms on single character recognition, few deals with the impact of the selected feature on segmentation. For recognition-based segmentation, a good recognition performance on single character may not always have good performance on segmentation. In this paper, two types of grayscale feature, the R-Feature and the S-Feature, are proposed based on dual-eigenspace decomposition. The RFeature is suitable for single character recognition. The SFeature is suitable for text string segmentation. These two feature are combined to further improve the performance for degraded Japanese text string recognition.

UR - http://www.scopus.com/inward/record.url?scp=34047194041&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34047194041&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:34047194041

SP - 39

EP - 44

T2 - 1st International Workshop on Camera-Based Document Analysis and Recognition, CBDAR 2005

Y2 - 29 August 2005 through 29 August 2005

ER -