Presentation | 1998/2/19 Color Segmentation to Extract Character Area from a Color Document Hiroyuki Hase, Hiroshi Maruyama, Yasuyuki Matsushita, Masaaki Yoneda, Mitsuru Sakai, |
---|---|
PDF Download Page | PDF download Page Link |
Abstract(in Japanese) | (See Japanese page) |
Abstract(in English) | It is very important that OCR systems can understand the character strings of a color document such as book covers or posters. In order to do this, we aim to extract character areas from a color document as the first step. Though we can see that the characters are of a single color, we can also recognize their color distribution by measurement. Consequently, we consider the desirable condition of color segmentation by clustering. However, too many clusters result in over-segmentation of a character color, while fewer clusters result in fusion with the background color. In this paper, we consider a segmentation method for the adequate extraction of character areas. We first observed the color distribution of a character area. And we also examined the extraction rate of a character area for the number of clusters. But the clustering is impractical for real time processing. So, we propose a simple algorithm for the extraction of representative colors. By using the algorithm, we can obtain adequate representative colors that depend on the input color image. Finally, we compared our method with the clustering method. |
Keyword(in Japanese) | (See Japanese page) |
Keyword(in English) | color document / character area / clustering / segmentation / representative color |
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) | Color Segmentation to Extract Character Area from a Color Document |
Sub Title (in English) | |
Keyword(1) | color document |
Keyword(2) | character area |
Keyword(3) | clustering |
Keyword(4) | segmentation |
Keyword(5) | representative color |
1st Author's Name | Hiroyuki Hase |
1st Author's Affiliation | Intellectual Information Systems Engineering, Faculty of Engineering, Toyama University() |
2nd Author's Name | Hiroshi Maruyama |
2nd Author's Affiliation | Intellectual Information Systems Engineering, Faculty of Engineering, Toyama University |
3rd Author's Name | Yasuyuki Matsushita |
3rd Author's Affiliation | Intellectual Information Systems Engineering, Faculty of Engineering, Toyama University |
4th Author's Name | Masaaki Yoneda |
4th Author's Affiliation | Intellectual Information Systems Engineering, Faculty of Engineering, Toyama University |
5th Author's Name | Mitsuru Sakai |
5th Author's Affiliation | Intellectual Information Systems Engineering, Faculty of Engineering, Toyama University |
Date | 1998/2/19 |
Paper # | |
Volume (vol) | vol.97 |
Number (no) | 558 |
Page | pp.pp.- |
#Pages | 8 |
Date of Issue |