Japanese scene character recognition using random image feature and ensemble scheme

Fuma Horie; Hideaki Goto

doi:10.5220/0007341904140420

Japanese scene character recognition using random image feature and ensemble scheme

Fuma Horie, Hideaki Goto

Research and Development Divisions

Tohoku University

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

1 Citation (Scopus)

Abstract

Scene character recognition is challenging and difficult owing to various environmental factors at image capturing and complex design of characters. Japanese character recognition requires a large number of scene character images for training since thousands of character classes exist in the language. In order to enhance the Japanese scene character recognition, we utilized a data augmentation method and an ensemble scheme in our previous work. In this paper, Random Image Feature (RI-Feature) method is newly proposed for improving the ensemble learning. Experimental results show that the accuracy has been improved from 65.57% to 78.50% by adding the RI-Feature method to the ensemble learning. It is also shown that HOG feature outperforms CNN in the Japanese scene character recognition.

Original language	English
Title of host publication	ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods
Editors	Maria De Marsico, Gabriella Sanniti di Baja, Ana Fred
Publisher	SciTePress
Pages	414-420
Number of pages	7
ISBN (Electronic)	9789897583513
DOIs	https://doi.org/10.5220/0007341904140420
Publication status	Published - 2019
Event	8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019 - Prague, Czech Republic Duration: 2019 Feb 19 → 2019 Feb 21

Publication series

Name	ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods

Conference

Conference	8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019
Country/Territory	Czech Republic
City	Prague
Period	19/2/19 → 19/2/21

Keywords

Ensemble Voting Classifier
Japanese Scene Character Recognition
Multi-Layer Perceptron
Random Image Feature
Synthetic Scene Character Data

Access to Document

10.5220/0007341904140420

Cite this

Horie, F., & Goto, H. (2019). Japanese scene character recognition using random image feature and ensemble scheme. In M. De Marsico, G. S. di Baja, & A. Fred (Eds.), ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods (pp. 414-420). (ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods). SciTePress. https://doi.org/10.5220/0007341904140420

Horie, Fuma ; Goto, Hideaki. / Japanese scene character recognition using random image feature and ensemble scheme. ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods. editor / Maria De Marsico ; Gabriella Sanniti di Baja ; Ana Fred. SciTePress, 2019. pp. 414-420 (ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods).

@inproceedings{1ce96645d3744f7fa3cd8580bc3813df,

title = "Japanese scene character recognition using random image feature and ensemble scheme",

abstract = "Scene character recognition is challenging and difficult owing to various environmental factors at image capturing and complex design of characters. Japanese character recognition requires a large number of scene character images for training since thousands of character classes exist in the language. In order to enhance the Japanese scene character recognition, we utilized a data augmentation method and an ensemble scheme in our previous work. In this paper, Random Image Feature (RI-Feature) method is newly proposed for improving the ensemble learning. Experimental results show that the accuracy has been improved from 65.57% to 78.50% by adding the RI-Feature method to the ensemble learning. It is also shown that HOG feature outperforms CNN in the Japanese scene character recognition.",

keywords = "Ensemble Voting Classifier, Japanese Scene Character Recognition, Multi-Layer Perceptron, Random Image Feature, Synthetic Scene Character Data",

author = "Fuma Horie and Hideaki Goto",

note = "Publisher Copyright: Copyright {\textcopyright} 2019 by SCITEPRESS - Science and Technology Publications, Lda. All rights reserved; 8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019 ; Conference date: 19-02-2019 Through 21-02-2019",

year = "2019",

doi = "10.5220/0007341904140420",

language = "English",

series = "ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods",

publisher = "SciTePress",

pages = "414--420",

editor = "{De Marsico}, Maria and {di Baja}, {Gabriella Sanniti} and Ana Fred",

booktitle = "ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods",

}

Horie, F & Goto, H 2019, Japanese scene character recognition using random image feature and ensemble scheme. in M De Marsico, GS di Baja & A Fred (eds), ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods. ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods, SciTePress, pp. 414-420, 8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019, Prague, Czech Republic, 19/2/19. https://doi.org/10.5220/0007341904140420

Japanese scene character recognition using random image feature and ensemble scheme. / Horie, Fuma; Goto, Hideaki.
ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods. ed. / Maria De Marsico; Gabriella Sanniti di Baja; Ana Fred. SciTePress, 2019. p. 414-420 (ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods).

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

TY - GEN

T1 - Japanese scene character recognition using random image feature and ensemble scheme

AU - Horie, Fuma

AU - Goto, Hideaki

PY - 2019

Y1 - 2019

N2 - Scene character recognition is challenging and difficult owing to various environmental factors at image capturing and complex design of characters. Japanese character recognition requires a large number of scene character images for training since thousands of character classes exist in the language. In order to enhance the Japanese scene character recognition, we utilized a data augmentation method and an ensemble scheme in our previous work. In this paper, Random Image Feature (RI-Feature) method is newly proposed for improving the ensemble learning. Experimental results show that the accuracy has been improved from 65.57% to 78.50% by adding the RI-Feature method to the ensemble learning. It is also shown that HOG feature outperforms CNN in the Japanese scene character recognition.

AB - Scene character recognition is challenging and difficult owing to various environmental factors at image capturing and complex design of characters. Japanese character recognition requires a large number of scene character images for training since thousands of character classes exist in the language. In order to enhance the Japanese scene character recognition, we utilized a data augmentation method and an ensemble scheme in our previous work. In this paper, Random Image Feature (RI-Feature) method is newly proposed for improving the ensemble learning. Experimental results show that the accuracy has been improved from 65.57% to 78.50% by adding the RI-Feature method to the ensemble learning. It is also shown that HOG feature outperforms CNN in the Japanese scene character recognition.

KW - Ensemble Voting Classifier

KW - Japanese Scene Character Recognition

KW - Multi-Layer Perceptron

KW - Random Image Feature

KW - Synthetic Scene Character Data

UR - http://www.scopus.com/inward/record.url?scp=85064634768&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85064634768&partnerID=8YFLogxK

U2 - 10.5220/0007341904140420

DO - 10.5220/0007341904140420

M3 - Conference contribution

AN - SCOPUS:85064634768

T3 - ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods

SP - 414

EP - 420

BT - ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods

A2 - De Marsico, Maria

A2 - di Baja, Gabriella Sanniti

A2 - Fred, Ana

PB - SciTePress

T2 - 8th International Conference on Pattern Recognition Applications and Methods, ICPRAM 2019

Y2 - 19 February 2019 through 21 February 2019

ER -

Horie F, Goto H. Japanese scene character recognition using random image feature and ensemble scheme. In De Marsico M, di Baja GS, Fred A, editors, ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods. SciTePress. 2019. p. 414-420. (ICPRAM 2019 - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods). doi: 10.5220/0007341904140420

Japanese scene character recognition using random image feature and ensemble scheme

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this