Share Email Print

Proceedings Paper

Gaussian process style transfer mapping for historical Chinese character recognition
Author(s): Jixiong Feng; Liangrui Peng; Franck Lebourgeois
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Historical Chinese character recognition is very important to larger scale historical document digitalization, but is a very challenging problem due to lack of labeled training samples. This paper proposes a novel non-linear transfer learning method, namely Gaussian Process Style Transfer Mapping (GP-STM). The GP-STM extends traditional linear Style Transfer Mapping (STM) by using Gaussian process and kernel methods. With GP-STM, existing printed Chinese character samples are used to help the recognition of historical Chinese characters. To demonstrate this framework, we compare feature extraction methods, train a modified quadratic discriminant function (MQDF) classifier on printed Chinese character samples, and implement the GP-STM model on Dunhuang historical documents. Various kernels and parameters are explored, and the impact of the number of training samples is evaluated. Experimental results show that accuracy increases by nearly 15 percentage points (from 42.8% to 57.5%) using GP-STM, with an improvement of more than 8 percentage points (from 49.2% to 57.5%) compared to the STM approach.

Paper Details

Date Published: 14 January 2015
PDF: 12 pages
Proc. SPIE 9402, Document Recognition and Retrieval XXII, 94020D (14 January 2015); doi: 10.1117/12.2076119
Show Author Affiliations
Jixiong Feng, Tsinghua Univ. (China)
Liangrui Peng, Tsinghua Univ. (China)
Franck Lebourgeois, Institut National des Sciences Appliquées de Lyon (France)

Published in SPIE Proceedings Vol. 9402:
Document Recognition and Retrieval XXII
Eric K. Ringger; Bart Lamiroy, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?