Presentation | 1997/10/20 Unconstrained Handwritten Numerals Recognition Using Line Features Extracted by DOG Filters Makoto SASAKI, Yuzo HIRAI, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | In pattern recognition problems, the type of features to be used is equally important as the type of classifieIrs. In this paper, it is shown that line features in eight orientations extracted by orientational DOG (Difference-OF-two-Gaussian) filters give a very high recognition rate in unconstrained handwritten numerals recognition with a standard back-propagation network. The unconstrained handwritten numerals database ETL-1 developed by Electro-Technical Laboratory of Japan was used to test the performance of the network. In our experiments, the recognition rate as high as 99.1% could be achieved for samples in a test set. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | line features extraction / orientational DOG filters / back propagation / handwritten numerals recognition |
Paper # | NC97-37 |
Date of Issue |
Conference Information | |
Committee | NC |
---|---|
Conference Date | 1997/10/20(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 | Neurocomputing (NC) |
---|---|
Language | JPN |
Title (in Japanese) | (See Japanese page) |
Sub Title (in Japanese) | (See Japanese page) |
Title (in English) | Unconstrained Handwritten Numerals Recognition Using Line Features Extracted by DOG Filters |
Sub Title (in English) | |
Keyword(1) | line features extraction |
Keyword(2) | orientational DOG filters |
Keyword(3) | back propagation |
Keyword(4) | handwritten numerals recognition |
1st Author's Name | Makoto SASAKI |
1st Author's Affiliation | Master Program of Science and Engineering, University of Tsukuba() |
2nd Author's Name | Yuzo HIRAI |
2nd Author's Affiliation | Institute of Information Sciences and Electronics, University of Tsukuba |
Date | 1997/10/20 |
Paper # | NC97-37 |
Volume (vol) | vol.97 |
Number (no) | 332 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |