Fast and accurate tree-based clustering for Japanese/Chinese character recognition

Yuichi Abe; Takahiro Sasaki; Hideaki Goto

doi:10.1007/978-3-642-41184-7_47

Fast and accurate tree-based clustering for Japanese/Chinese character recognition

Yuichi Abe, Takahiro Sasaki, Hideaki Goto

Research and Development Divisions

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

2 Citations (Scopus)

Abstract

Recognizing text in natural scene images is very important to develop various systems such as an assistant device for visually-impaired people. Multilingual scene text recognition is also becoming important for wearable camera devices with language translation feature. Since computational resources are limited on such mobile devices, fast and accurate Optical Character Recognition (OCR) algorithm is needed. Nearest Neighbor (NN) search is quite popular in feature vector-based OCR systems, and its speed improvement is required. In this paper, we develop an OCR scheme with tree-based clustering technique with LDA (Linear Discriminant Analysis) aiming at real-time Japanese/Chinese character recognition. The experimental results using ETL9B dataset show that our proposed method is 94.6% faster than our previous method, also beating other techniques, at mere 0.24% accuracy drop from the full linear search.

Original language	English
Title of host publication	Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings
Pages	459-468
Number of pages	10
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-41184-7_47
Publication status	Published - 2013
Event	17th International Conference on Image Analysis and Processing, ICIAP 2013 - Naples, Italy Duration: 2013 Sept 9 → 2013 Sept 13

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	8157 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Other

Other	17th International Conference on Image Analysis and Processing, ICIAP 2013
Country/Territory	Italy
City	Naples
Period	13/9/9 → 13/9/13

Keywords

Approximate Nearest Neighbor (ANN) search
Fast Nearest Neighbor search
Linear Discriminant Analysis (LDA)
multilingual OCR
real-time character recognition

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-642-41184-7_47

Cite this

Abe, Y., Sasaki, T., & Goto, H. (2013). Fast and accurate tree-based clustering for Japanese/Chinese character recognition. In Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings (PART 2 ed., pp. 459-468). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8157 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-41184-7_47

Fast and accurate tree-based clustering for Japanese/Chinese character recognition. / Abe, Yuichi; Sasaki, Takahiro; Goto, Hideaki.
Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings. PART 2. ed. 2013. p. 459-468 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8157 LNCS, No. PART 2).

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

Abe, Y, Sasaki, T & Goto, H 2013, Fast and accurate tree-based clustering for Japanese/Chinese character recognition. in Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 8157 LNCS, pp. 459-468, 17th International Conference on Image Analysis and Processing, ICIAP 2013, Naples, Italy, 13/9/9. https://doi.org/10.1007/978-3-642-41184-7_47

Abe Y, Sasaki T, Goto H. Fast and accurate tree-based clustering for Japanese/Chinese character recognition. In Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings. PART 2 ed. 2013. p. 459-468. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-41184-7_47

Abe, Yuichi ; Sasaki, Takahiro ; Goto, Hideaki. / Fast and accurate tree-based clustering for Japanese/Chinese character recognition. Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings. PART 2. ed. 2013. pp. 459-468 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{da3e67a15c3a4b219c380ba7418f9051,

title = "Fast and accurate tree-based clustering for Japanese/Chinese character recognition",

abstract = "Recognizing text in natural scene images is very important to develop various systems such as an assistant device for visually-impaired people. Multilingual scene text recognition is also becoming important for wearable camera devices with language translation feature. Since computational resources are limited on such mobile devices, fast and accurate Optical Character Recognition (OCR) algorithm is needed. Nearest Neighbor (NN) search is quite popular in feature vector-based OCR systems, and its speed improvement is required. In this paper, we develop an OCR scheme with tree-based clustering technique with LDA (Linear Discriminant Analysis) aiming at real-time Japanese/Chinese character recognition. The experimental results using ETL9B dataset show that our proposed method is 94.6% faster than our previous method, also beating other techniques, at mere 0.24% accuracy drop from the full linear search.",

keywords = "Approximate Nearest Neighbor (ANN) search, Fast Nearest Neighbor search, Linear Discriminant Analysis (LDA), multilingual OCR, real-time character recognition",

author = "Yuichi Abe and Takahiro Sasaki and Hideaki Goto",

year = "2013",

doi = "10.1007/978-3-642-41184-7_47",

language = "English",

isbn = "9783642411830",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "459--468",

booktitle = "Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings",

edition = "PART 2",

note = "17th International Conference on Image Analysis and Processing, ICIAP 2013 ; Conference date: 09-09-2013 Through 13-09-2013",

}

TY - GEN

T1 - Fast and accurate tree-based clustering for Japanese/Chinese character recognition

AU - Abe, Yuichi

AU - Sasaki, Takahiro

AU - Goto, Hideaki

PY - 2013

Y1 - 2013

N2 - Recognizing text in natural scene images is very important to develop various systems such as an assistant device for visually-impaired people. Multilingual scene text recognition is also becoming important for wearable camera devices with language translation feature. Since computational resources are limited on such mobile devices, fast and accurate Optical Character Recognition (OCR) algorithm is needed. Nearest Neighbor (NN) search is quite popular in feature vector-based OCR systems, and its speed improvement is required. In this paper, we develop an OCR scheme with tree-based clustering technique with LDA (Linear Discriminant Analysis) aiming at real-time Japanese/Chinese character recognition. The experimental results using ETL9B dataset show that our proposed method is 94.6% faster than our previous method, also beating other techniques, at mere 0.24% accuracy drop from the full linear search.

AB - Recognizing text in natural scene images is very important to develop various systems such as an assistant device for visually-impaired people. Multilingual scene text recognition is also becoming important for wearable camera devices with language translation feature. Since computational resources are limited on such mobile devices, fast and accurate Optical Character Recognition (OCR) algorithm is needed. Nearest Neighbor (NN) search is quite popular in feature vector-based OCR systems, and its speed improvement is required. In this paper, we develop an OCR scheme with tree-based clustering technique with LDA (Linear Discriminant Analysis) aiming at real-time Japanese/Chinese character recognition. The experimental results using ETL9B dataset show that our proposed method is 94.6% faster than our previous method, also beating other techniques, at mere 0.24% accuracy drop from the full linear search.

KW - Approximate Nearest Neighbor (ANN) search

KW - Fast Nearest Neighbor search

KW - Linear Discriminant Analysis (LDA)

KW - multilingual OCR

KW - real-time character recognition

UR - http://www.scopus.com/inward/record.url?scp=84884713881&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884713881&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-41184-7_47

DO - 10.1007/978-3-642-41184-7_47

M3 - Conference contribution

AN - SCOPUS:84884713881

SN - 9783642411830

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

SP - 459

EP - 468

BT - Image Analysis and Processing, ICIAP 2013 - 17th International Conference, Proceedings

T2 - 17th International Conference on Image Analysis and Processing, ICIAP 2013

Y2 - 9 September 2013 through 13 September 2013

ER -

Fast and accurate tree-based clustering for Japanese/Chinese character recognition

Abstract

Publication series

Other

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this