Presentation 1998/2/19
An Efficiency Improvement of the Line Pattern Parsing Algorithm Based on Two-dimensional CYK Method
Naoto KAWASHITA, Yoshinori KATAYAMA, Hiroaki SAKOE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We proposed a CYK based two-dimensional parsing method for directly detecting a line pattern in a noisy picture according to a context free grammar (CFG) model. The time and spatial complexities of its basic algorithm amount to Ο(Ν^6) and Ο(ΜΝ^4), respectively, for Ν×Ν input image and Μ symbols Model. In this paper we investigate several improvements for efficiency enhancements. They include (1) the analyzed data limitation by removing background area from the image, (2) the symbol driven limitations for line tracing and rule application, (3) the shared parsing table among symbols. Significant efficiency improvement of the proposed algorithm was observed through experiments.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Line pattern analysis / CFG modeling / Two-dimensional CYK parsing
Paper #
Date of Issue

Conference Information
Committee PRMU
Conference Date 1998/2/19(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Pattern Recognition and Media Understanding (PRMU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) An Efficiency Improvement of the Line Pattern Parsing Algorithm Based on Two-dimensional CYK Method
Sub Title (in English)
Keyword(1) Line pattern analysis
Keyword(2) CFG modeling
Keyword(3) Two-dimensional CYK parsing
1st Author's Name Naoto KAWASHITA
1st Author's Affiliation Nishinippon Shimbun()
2nd Author's Name Yoshinori KATAYAMA
2nd Author's Affiliation Graduate School of Information Science and Electrical Engineering, Kyushu University
3rd Author's Name Hiroaki SAKOE
3rd Author's Affiliation Graduate School of Information Science and Electrical Engineering, Kyushu University
Date 1998/2/19
Paper #
Volume (vol) vol.97
Number (no) 558
Page pp.pp.-
#Pages 8
Date of Issue