Chaincode contour processing for handwritten word recognition

被引：37

作者：

Madhvanath, S

Kim, G

Govindaraju, V

机构：

[1] IBM Corp, San Jose, CA 95120 USA

[2] Sogang Univ, Dept Elect Engn, Seoul, South Korea

[3] SUNY Buffalo, Dept Comp Sci, CEDAR, Buffalo, NY 14228 USA

来源：

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE | 1999年 / 21卷 / 09期

关键词：

image processing; chain code; handwriting recognition; preprocessing; segmentation; feature extraction;

D O I：

10.1109/34.790433

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Contour representations of binary images of handwritten words afford considerable reduction in storage requirements while providing lossless representation. On the other hand, the one-dimensional nature of contours presents interesting challenges for processing images for handwritten word recognition. Our experiments indicate that significant gains are to be realized in both speed and recognition accuracy by using a contour representation in handwriting applications.

引用

页码：928 / 932

页数：5

共 13 条

[1] SHAPE DESCRIPTION USING WEIGHTED SYMMETRIC AXIS FEATURES [J].

BLUM, H ;

NAGEL, RN .

PATTERN RECOGNITION, 1978, 10 (03) :167-180

[2] PREPROCESSING TECHNIQUES FOR CURSIVE SCRIPT WORD RECOGNITION [J].

BROWN, MK ;

GANAPATHY, S .

PATTERN RECOGNITION, 1983, 16 (05) :447-458

[3]

Caesar T., 1995, Proceedings of the Third International Conference on Document Analysis and Recognition, P382, DOI 10.1109/ICDAR.1995.599018

[4] Automatic reading of cursive scripts using a reading model and perceptual concepts: The PERPECTO system [J].

Côté M. ;

Lecolinet E. ;

Cheriet M. ;

Suen C.Y. .

International Journal on Document Analysis and Recognition, 1998, 1 (1) :3-17

[5]

Freeman H., 1974, Computing Surveys, V6, P57, DOI 10.1145/356625.356627

[6]

HOLT MJJ, 1992, PIXELS FEATURES, V2, P41

[7] A lexicon driven approach to handwritten word recognition for real-time applications [J].

Kim, G ;

Govindaraju, V .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (04) :366-379

[8] Efficient chain code based image manipulation for handwritten word recognition [J].

Kim, GH ;

Govindaraju, V .

DOCUMENT RECOGNITION III, 1996, 2660 :262-272

[9]

Kimura F., 1993, Proceedings of the Second International Conference on Document Analysis and Recognition (Cat. No.93TH0578-5), P18, DOI 10.1109/ICDAR.1993.395791

[10]

KIMURA F, 1993, P 3 INT WORKSH FRONT, P122

← 1 2 →