A further step to perfect accuracy by training CNN with larger data

Seiichi Uchida, Shota Ide, Brian Kenji Iwana, Anna Zhu

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

7 Citations (Scopus)

Abstract

Convolutional Neural Networks (CNN) are on the forefront of accurate character recognition. This paper explores CNNs at their maximum capacity by implementing the use of large datasets. We show a near-perfect performance by using a dataset of about 820,000 real samples of isolated handwritten digits, much larger than the conventional MNIST database. In addition, we report a near-perfect performance on the recognition of machine-printed digits and multi-font digital born digits. Also, in order to progress toward a universal OCR, we propose methods of combining the datasets into one classifier. This paper reveals the effects of combining the datasets prior to training and the effects of transfer learning during training. The results of the proposed methods also show an almost perfect accuracy suggesting the ability of the network to generalize all forms of text.

Original languageEnglish
Title of host publicationProceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages405-410
Number of pages6
ISBN (Electronic)9781509009817
DOIs
Publication statusPublished - Jul 2 2016
Event15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016 - Shenzhen, China
Duration: Oct 23 2016Oct 26 2016

Publication series

NameProceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR
Volume0
ISSN (Print)2167-6445
ISSN (Electronic)2167-6453

Other

Other15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016
CountryChina
CityShenzhen
Period10/23/1610/26/16

Fingerprint

Optical character recognition
Character recognition
Classifiers
Neural networks

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Cite this

Uchida, S., Ide, S., Iwana, B. K., & Zhu, A. (2016). A further step to perfect accuracy by training CNN with larger data. In Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016 (pp. 405-410). [7814098] (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR; Vol. 0). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICFHR.2016.0082

A further step to perfect accuracy by training CNN with larger data. / Uchida, Seiichi; Ide, Shota; Iwana, Brian Kenji; Zhu, Anna.

Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016. Institute of Electrical and Electronics Engineers Inc., 2016. p. 405-410 7814098 (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR; Vol. 0).

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

Uchida, S, Ide, S, Iwana, BK & Zhu, A 2016, A further step to perfect accuracy by training CNN with larger data. in Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016., 7814098, Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR, vol. 0, Institute of Electrical and Electronics Engineers Inc., pp. 405-410, 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016, Shenzhen, China, 10/23/16. https://doi.org/10.1109/ICFHR.2016.0082
Uchida S, Ide S, Iwana BK, Zhu A. A further step to perfect accuracy by training CNN with larger data. In Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016. Institute of Electrical and Electronics Engineers Inc. 2016. p. 405-410. 7814098. (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR). https://doi.org/10.1109/ICFHR.2016.0082
Uchida, Seiichi ; Ide, Shota ; Iwana, Brian Kenji ; Zhu, Anna. / A further step to perfect accuracy by training CNN with larger data. Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016. Institute of Electrical and Electronics Engineers Inc., 2016. pp. 405-410 (Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR).
@inproceedings{e184276cfa1a4a5495bd54a6eee624be,
title = "A further step to perfect accuracy by training CNN with larger data",
abstract = "Convolutional Neural Networks (CNN) are on the forefront of accurate character recognition. This paper explores CNNs at their maximum capacity by implementing the use of large datasets. We show a near-perfect performance by using a dataset of about 820,000 real samples of isolated handwritten digits, much larger than the conventional MNIST database. In addition, we report a near-perfect performance on the recognition of machine-printed digits and multi-font digital born digits. Also, in order to progress toward a universal OCR, we propose methods of combining the datasets into one classifier. This paper reveals the effects of combining the datasets prior to training and the effects of transfer learning during training. The results of the proposed methods also show an almost perfect accuracy suggesting the ability of the network to generalize all forms of text.",
author = "Seiichi Uchida and Shota Ide and Iwana, {Brian Kenji} and Anna Zhu",
year = "2016",
month = "7",
day = "2",
doi = "10.1109/ICFHR.2016.0082",
language = "English",
series = "Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "405--410",
booktitle = "Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016",
address = "United States",

}

TY - GEN

T1 - A further step to perfect accuracy by training CNN with larger data

AU - Uchida, Seiichi

AU - Ide, Shota

AU - Iwana, Brian Kenji

AU - Zhu, Anna

PY - 2016/7/2

Y1 - 2016/7/2

N2 - Convolutional Neural Networks (CNN) are on the forefront of accurate character recognition. This paper explores CNNs at their maximum capacity by implementing the use of large datasets. We show a near-perfect performance by using a dataset of about 820,000 real samples of isolated handwritten digits, much larger than the conventional MNIST database. In addition, we report a near-perfect performance on the recognition of machine-printed digits and multi-font digital born digits. Also, in order to progress toward a universal OCR, we propose methods of combining the datasets into one classifier. This paper reveals the effects of combining the datasets prior to training and the effects of transfer learning during training. The results of the proposed methods also show an almost perfect accuracy suggesting the ability of the network to generalize all forms of text.

AB - Convolutional Neural Networks (CNN) are on the forefront of accurate character recognition. This paper explores CNNs at their maximum capacity by implementing the use of large datasets. We show a near-perfect performance by using a dataset of about 820,000 real samples of isolated handwritten digits, much larger than the conventional MNIST database. In addition, we report a near-perfect performance on the recognition of machine-printed digits and multi-font digital born digits. Also, in order to progress toward a universal OCR, we propose methods of combining the datasets into one classifier. This paper reveals the effects of combining the datasets prior to training and the effects of transfer learning during training. The results of the proposed methods also show an almost perfect accuracy suggesting the ability of the network to generalize all forms of text.

UR - http://www.scopus.com/inward/record.url?scp=85012874404&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85012874404&partnerID=8YFLogxK

U2 - 10.1109/ICFHR.2016.0082

DO - 10.1109/ICFHR.2016.0082

M3 - Conference contribution

AN - SCOPUS:85012874404

T3 - Proceedings of International Conference on Frontiers in Handwriting Recognition, ICFHR

SP - 405

EP - 410

BT - Proceedings - 2016 15th International Conference on Frontiers in Handwriting Recognition, ICFHR 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -