Presentation 1995/9/28
Region Extraction and Recognition of Characters with Graphical Backgrounds based on Complementary Similarity Measure
Minako Sawaki, Norihiro Hagita,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This paper proposes a method for recognizing characters with graphical backgrounds in Japanese newspaper headlines using the complementary similarity measure. First, text line regions are extracted from input images using the projection profile based on the complementary similarity measure. Next, the window in the text-line is displaced continuously along the line with matching it against reference patterns of character categories. The specified category is determined as a recognition category when the maximum complementary similarity exceeds a threshold. Experiments show that this method achieves a high recognition rate of 91.1% for 50 Japanese headline images.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) character recognition / 0CR / character segmentation / displacement matching
Paper # PRU95-106
Date of Issue

Conference Information
Committee PRU
Conference Date 1995/9/28(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 Understanding (PRU)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Region Extraction and Recognition of Characters with Graphical Backgrounds based on Complementary Similarity Measure
Sub Title (in English)
Keyword(1) character recognition
Keyword(2) 0CR
Keyword(3) character segmentation
Keyword(4) displacement matching
1st Author's Name Minako Sawaki
1st Author's Affiliation NTT Basic Research Laboratories()
2nd Author's Name Norihiro Hagita
2nd Author's Affiliation NTT Basic Research Laboratories
Date 1995/9/28
Paper # PRU95-106
Volume (vol) vol.95
Number (no) 278
Page pp.pp.-
#Pages 6
Date of Issue