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

15 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

All Science Journal Classification (ASJC) codes

  • Computer Science Applications
  • Computer Vision and Pattern Recognition

Fingerprint Dive into the research topics of 'A further step to perfect accuracy by training CNN with larger data'. Together they form a unique fingerprint.

  • 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