OS15-2 A method of end-to-end self-understanding of Chinese paper-dictionaries

Title:	OS15-2 A method of end-to-end self-understanding of Chinese paper-dictionaries
Publication:	ICAROB2018
Volume:	23
Pages:	578-581
ISSN:	2188-7829
DOI:	10.5954/ICAROB.2018.OS15-2
Author(s):	Zhijian Lv, Yizhun Peng
Publication Date:	February 2, 2018
Keywords:	end-to-end self-understanding, image-segmentation
Abstract:	This paper introduces a method of end-to-end self-understanding of Chinese paper-dictionaries. In this method, a page of Chinese paper-dictionaries is scanned into an electronic image. And then the electronic image is preprocessed, including un-distortion, side scrapping, binarization, and so on. Finally, using an end-to end deep learning method, the pre-processed electronic image is intelligently segmented, text recognized, and context understood. Our method has been applied to self-understand a serial of Chinese paper-dictionaries, which have more than 13000 pages and 2.1 millions of entries. And its correct rate of self-understanding of Chinese phrases is more than 99.5%. Its performance has proved it's availability.
PDF File:	https://alife-robotics.co.jp/members2018/icarob/data/html/data/OS_pdf/OS15/OS15-2.pdf
Copyright:	© The authors. This article is distributed under the terms of the Creative Commons Attribution License 4.0, which permits non-commercial use, distribution and reproduction in any medium, provided the original work is properly cited. See for details: https://creativecommons.org/licenses/by-nc/4.0/