Presentation 2000/2/22
Efficacy Evaluation of Directional Edge Component Projection Method for Caption Recognition
Hiroshi Yoshimura, Kenji Kondo, Minoru Etoh, Naokazu Yokoya,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) We propose a gray-scale character recognition method using Gabor filter. In our method, features are derived by accumulating edge components in local region of output of Gabor filters. This method is robust for charcter positioning errors and can recognize low-resolution characters precisely. The average of recognition rate is 88.8% for 865 alphabets obtained from seven films. We also experiment on characters with positioning errors. From the result, proposed method is robust for displacement less than 5% of character size. It shows that this method is useful for video caption recognition.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Character Recognition / gra-scale character image / Gabor Filter / edge feature / video indexing
Paper # PRMU99-233
Date of Issue

Conference Information
Committee PRMU
Conference Date 2000/2/22(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) Efficacy Evaluation of Directional Edge Component Projection Method for Caption Recognition
Sub Title (in English)
Keyword(1) Character Recognition
Keyword(2) gra-scale character image
Keyword(3) Gabor Filter
Keyword(4) edge feature
Keyword(5) video indexing
1st Author's Name Hiroshi Yoshimura
1st Author's Affiliation Nara Institute of Science and Technology(NAIST)()
2nd Author's Name Kenji Kondo
2nd Author's Affiliation Advanced Technology Research Laboratories, Matsusita Electric Industrial Co., Ltd.
3rd Author's Name Minoru Etoh
3rd Author's Affiliation Nara Institute of Science and Technology(NAIST):Information & Network Research Laboratory, Matsusita Research Institute Tokyo, Inc.
4th Author's Name Naokazu Yokoya
4th Author's Affiliation Nara Institute of Science and Technology(NAIST)
Date 2000/2/22
Paper # PRMU99-233
Volume (vol) vol.99
Number (no) 649
Page pp.pp.-
#Pages 6
Date of Issue