Abstract
We propose a method that extracts text regions in natural scene images using low-level image features and that verifies the extracted regions through a high-level text stroke feature. Then the two level features are combined hierarchically. The low-level features are color continuity, gray-level variation and color variance. The color continuity is used since most of the characters in a text region have the same color, and the gray-level variation is used since the text strokes are distinctive to the background in their gray-level values. Also, the color variance is used since the text strokes are distinctive in their colors to the background, and this value is more sensitive than the gray-level variations. As a high level feature, text stroke is examined using multi-resolution wavelet transforms on local image areas and the feature vector is input to a SVM(Support Vector Machine) for verification. We tested the proposed method with various kinds of the natural scene images and confirmed that extraction rates are high even in complex images.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004 |
| Editors | J. Kittler, M. Petrou, M. Nixon |
| Pages | 679-682 |
| Number of pages | 4 |
| DOIs | |
| Publication status | Published - 2004 Dec 17 |
| Event | Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004 - Cambridge, United Kingdom Duration: 2004 Aug 23 → 2004 Aug 26 |
Publication series
| Name | Proceedings - International Conference on Pattern Recognition |
|---|---|
| Volume | 2 |
| ISSN (Print) | 1051-4651 |
Other
| Other | Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004 |
|---|---|
| Country | United Kingdom |
| City | Cambridge |
| Period | 04/8/23 → 04/8/26 |
Fingerprint
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
Cite this
}
Scene text extraction in natural scene images using hierarchical feature combining and verification. / Kim, K. C.; Byun, H. R.; Song, Y. J.; Choi, Y. W.; Chi, S. Y.; Kim, K. K.; Chung, Y. K.
Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004. ed. / J. Kittler; M. Petrou; M. Nixon. 2004. p. 679-682 (Proceedings - International Conference on Pattern Recognition; Vol. 2).Research output: Chapter in Book/Report/Conference proceeding › Conference contribution
TY - GEN
T1 - Scene text extraction in natural scene images using hierarchical feature combining and verification
AU - Kim, K. C.
AU - Byun, H. R.
AU - Song, Y. J.
AU - Choi, Y. W.
AU - Chi, S. Y.
AU - Kim, K. K.
AU - Chung, Y. K.
PY - 2004/12/17
Y1 - 2004/12/17
N2 - We propose a method that extracts text regions in natural scene images using low-level image features and that verifies the extracted regions through a high-level text stroke feature. Then the two level features are combined hierarchically. The low-level features are color continuity, gray-level variation and color variance. The color continuity is used since most of the characters in a text region have the same color, and the gray-level variation is used since the text strokes are distinctive to the background in their gray-level values. Also, the color variance is used since the text strokes are distinctive in their colors to the background, and this value is more sensitive than the gray-level variations. As a high level feature, text stroke is examined using multi-resolution wavelet transforms on local image areas and the feature vector is input to a SVM(Support Vector Machine) for verification. We tested the proposed method with various kinds of the natural scene images and confirmed that extraction rates are high even in complex images.
AB - We propose a method that extracts text regions in natural scene images using low-level image features and that verifies the extracted regions through a high-level text stroke feature. Then the two level features are combined hierarchically. The low-level features are color continuity, gray-level variation and color variance. The color continuity is used since most of the characters in a text region have the same color, and the gray-level variation is used since the text strokes are distinctive to the background in their gray-level values. Also, the color variance is used since the text strokes are distinctive in their colors to the background, and this value is more sensitive than the gray-level variations. As a high level feature, text stroke is examined using multi-resolution wavelet transforms on local image areas and the feature vector is input to a SVM(Support Vector Machine) for verification. We tested the proposed method with various kinds of the natural scene images and confirmed that extraction rates are high even in complex images.
UR - http://www.scopus.com/inward/record.url?scp=10044240329&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=10044240329&partnerID=8YFLogxK
U2 - 10.1109/ICPR.2004.1334350
DO - 10.1109/ICPR.2004.1334350
M3 - Conference contribution
AN - SCOPUS:10044240329
SN - 0769521282
T3 - Proceedings - International Conference on Pattern Recognition
SP - 679
EP - 682
BT - Proceedings of the 17th International Conference on Pattern Recognition, ICPR 2004
A2 - Kittler, J.
A2 - Petrou, M.
A2 - Nixon, M.
ER -