Difference between revisions of "Publications/minetto.10.icip"
From LRDE
(Created page with "{{Publication | date = 2010-09-01 | authors = Rodrigo Minetto, Nicolas Thome, Matthieu Cord, Jonathan Fabrizio, Beatriz Marcotegui | title = SnooperText: A Multiresolution Sys...") |
|||
(7 intermediate revisions by the same user not shown) | |||
Line 1: | Line 1: | ||
{{Publication |
{{Publication |
||
− | | |
+ | | published = true |
+ | | date = 2010-12-31 |
||
| authors = Rodrigo Minetto, Nicolas Thome, Matthieu Cord, Jonathan Fabrizio, Beatriz Marcotegui |
| authors = Rodrigo Minetto, Nicolas Thome, Matthieu Cord, Jonathan Fabrizio, Beatriz Marcotegui |
||
| title = SnooperText: A Multiresolution System for Text Detection in Complex Visual Scenes |
| title = SnooperText: A Multiresolution System for Text Detection in Complex Visual Scenes |
||
Line 7: | Line 8: | ||
| address = Hong Kong |
| address = Hong Kong |
||
| abstract = Text detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good results |
| abstract = Text detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good results |
||
+ | | lrdeprojects = Olena |
||
+ | | lrdenewsdate = 2010-12-31 |
||
| type = inproceedings |
| type = inproceedings |
||
| id = minetto.10.icip |
| id = minetto.10.icip |
||
Line 13: | Line 16: | ||
author = <nowiki>{</nowiki>Rodrigo Minetto and Nicolas Thome and Matthieu Cord and |
author = <nowiki>{</nowiki>Rodrigo Minetto and Nicolas Thome and Matthieu Cord and |
||
Jonathan Fabrizio and Beatriz Marcotegui<nowiki>}</nowiki>, |
Jonathan Fabrizio and Beatriz Marcotegui<nowiki>}</nowiki>, |
||
− | title = <nowiki>{</nowiki> |
+ | title = <nowiki>{</nowiki>Snooper<nowiki>{</nowiki>T<nowiki>}</nowiki>ext: <nowiki>{</nowiki>A<nowiki>}</nowiki> Multiresolution System for Text |
− | in Complex Visual Scenes<nowiki>}</nowiki>, |
+ | Detection in Complex Visual Scenes<nowiki>}</nowiki>, |
booktitle = <nowiki>{</nowiki>Proceedings of the IEEE International Conference on Image |
booktitle = <nowiki>{</nowiki>Proceedings of the IEEE International Conference on Image |
||
Processing (ICIP)<nowiki>}</nowiki>, |
Processing (ICIP)<nowiki>}</nowiki>, |
Latest revision as of 11:58, 3 December 2022
- Authors
- Rodrigo Minetto, Nicolas Thome, Matthieu Cord, Jonathan Fabrizio, Beatriz Marcotegui
- Where
- Proceedings of the IEEE International Conference on Image Processing (ICIP)
- Place
- Hong Kong
- Type
- inproceedings
- Projects
- Olena
- Date
- 2010-12-31
Abstract
Text detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good results
Bibtex (lrde.bib)
@InProceedings{ minetto.10.icip, author = {Rodrigo Minetto and Nicolas Thome and Matthieu Cord and Jonathan Fabrizio and Beatriz Marcotegui}, title = {Snooper{T}ext: {A} Multiresolution System for Text Detection in Complex Visual Scenes}, booktitle = {Proceedings of the IEEE International Conference on Image Processing (ICIP)}, pages = {3861--3864}, year = 2010, address = {Hong Kong}, month = sep, abstract = {Text detection in natural images remains a very challenging task. For instance, in an urban context, the detection is very difficult due to large variations in terms of shape, size, color, orientation, and the image may be blurred or have irregular illumination, etc. In this paper, we describe a robust and accurate multiresolution approach to detect and classify text regions in such scenarios. Based on generation/validation paradigm, we first segment images to detect character regions with a multiresolution algorithm able to manage large character size variations. The segmented regions are then filtered out using shapebased classification, and neighboring characters are merged to generate text hypotheses. A validation step computes a region signature based on texture analysis to reject false positives. We evaluate our algorithm in two challenging databases, achieving very good results}, keywords = {Text detection, multiresolution, image segmentation, machine learning} }