GPU-Based Fast Training of Discriminative Learning Quadratic Discriminant Function for Handwritten Chinese Character Recognition

Author

Ming-Ke Zhou ; Fei Yin ; Cheng-Lin Liu

Author_Institution

Nat. Lab. of Pattern Recognition, Inst. of Autom., Beijing, China

fYear

2013

fDate

25-28 Aug. 2013

Firstpage

842

Lastpage

846

Abstract

The discriminative training of classifiers for handwritten Chinese character recognition (HCCR) is highly demanding in computation due to the large number of categories. The inability of discriminative training with large sample set on personal computers has hindered the accuracy promotion for HCCR. To overcome this problem, we have implemented the training algorithm of discriminative learning quadratic discriminant function (DLQDF) on our graphics processing units (GPU) server, and have achieved 15 times speedup compared to single-core computation. By enlarging training sample set via distortion on a standard dataset of 3,755 classes, we could train the DLQDF on more than 50 million samples within 150min and get the test accuracy improved by 1.36%.

Keywords

graphics processing units; handwritten character recognition; image classification; learning (artificial intelligence); natural language processing; DLQDF; GPU-based fast training sample set; HCCR; discriminative learning quadratic discriminant function; discriminative training; graphics processing unit server; handwritten Chinese character recognition; personal computers; Acceleration; Accuracy; Character recognition; Feature extraction; Graphics processing units; Training; Vectors; GPU parallel computing; discriminative learning; handwritten Chinese character recognition; modified quadratic discriminant function; sample synthesis;

fLanguage

English

Publisher

ieee

Conference_Titel

Document Analysis and Recognition (ICDAR), 2013 12th International Conference on

Conference_Location

Washington, DC

ISSN

1520-5363

Type

conf

DOI

10.1109/ICDAR.2013.172

Filename

6628737