| Presentation | 2000/10/12 Extraction of text regions from scenery images based on multiple features Hiroki Takahashi, Kouichi Kasai, Masayuki Nakajima, |
|---|---|
| PDF Download Page | PDF download Page Link |
| Abstract(in Japanese) | (See Japanese page) |
| Abstract(in English) | This paper describes a method that extracts text regions from scenery images taken by digital camera. Many methods have been proposed for extraction of text regions from images. Most of them use a specific feature such as edge feature, color feature and so on. They, therefore, employ many conditions for extraction and it is also difficult to assign these parameters. This paper proposes an extraction method of text regions, which employs simple conditions, by using two kinds of features. One of them is edge feature which reflects local aspect of an image domain. Closed edges in an edge image are utilized for the extraction. The other feature is color distribution which reflects global aspect. Color clustering is employed for segmentation of images. In each feature image, candidates of character regions are selected based on relative geometric data among regions. Common regions of the selected one are extracted as character regions. Finally, these extracted regions are reconstructed as text regions. At the end of this paper, extracted text images from several scenery images and results of character recognition using a commercial software are shown. |
| Keyword(in Japanese) | (See Japanese page) |
| Keyword(in English) | |
| Paper # | HCS2000-30 |
| Date of Issue |
| Conference Information | |
| Committee | HCS |
|---|---|
| Conference Date | 2000/10/12(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 | Human Communication Science (HCS) |
|---|---|
| Language | JPN |
| Title (in Japanese) | (See Japanese page) |
| Sub Title (in Japanese) | (See Japanese page) |
| Title (in English) | Extraction of text regions from scenery images based on multiple features |
| Sub Title (in English) | |
| Keyword(1) | |
| 1st Author's Name | Hiroki Takahashi |
| 1st Author's Affiliation | Graduate School of Information Science & Engineering, Tokyo Institute of Technology() |
| 2nd Author's Name | Kouichi Kasai |
| 2nd Author's Affiliation | Graduate School of Information Science & Engineering, Tokyo Institute of Technology |
| 3rd Author's Name | Masayuki Nakajima |
| 3rd Author's Affiliation | Graduate School of Information Science & Engineering, Tokyo Institute of Technology |
| Date | 2000/10/12 |
| Paper # | HCS2000-30 |
| Volume (vol) | vol.100 |
| Number (no) | 375 |
| Page | pp.pp.- |
| #Pages | 8 |
| Date of Issue | |