|
| This Article | ||
| ||
| Share | ||
| Bibliographic References | ||
| Add to: | ||
 Digg  Furl  Spurl  Blink  Simpy  Google  Del.icio.us  Y!MyWeb | ||
| Search | ||
| ||
2011 International Conference on Document Analysis and Recognition
Minimizing User Annotations in the Generation of Layout Ground-Truthed Data
Beijing, China
September 18-September 21
ISBN: 978-0-7695-4520-2
| ASCII Text | x | ||
| Karim Hadjar, Rolf Ingold, "Minimizing User Annotations in the Generation of Layout Ground-Truthed Data," Document Analysis and Recognition, International Conference on, pp. 703-707, 2011 International Conference on Document Analysis and Recognition, 2011. | |||
| BibTex | x | ||
| @article{ 10.1109/ICDAR.2011.147, author = {Karim Hadjar and Rolf Ingold}, title = {Minimizing User Annotations in the Generation of Layout Ground-Truthed Data}, journal ={Document Analysis and Recognition, International Conference on}, volume = {0}, year = {2011}, issn = {1520-5363}, pages = {703-707}, doi = {http://doi.ieeecomputersociety.org/10.1109/ICDAR.2011.147}, publisher = {IEEE Computer Society}, address = {Los Alamitos, CA, USA}, } | |||
| RefWorks Procite/RefMan/Endnote | x | ||
| TY - CONF JO - Document Analysis and Recognition, International Conference on TI - Minimizing User Annotations in the Generation of Layout Ground-Truthed Data SN - 1520-5363 SP703 EP707 A1 - Karim Hadjar, A1 - Rolf Ingold, PY - 2011 KW - Ground Truth KW - Physical Layout Extraction KW - Datasets KW - Document Image KW - Artificial Neural Networks KW - Arabic Newspapers VL - 0 JA - Document Analysis and Recognition, International Conference on ER - | |||
This paper describes the adaptation of a previously developed document recognition framework called PLANET (Physical Layout Analysis of complex structured Arabic documents using artificial neural NETs) into a ground truthing system for complex Arabic document images [8]. PLANET is a layout analysis tool for Arabic documents with complex structures allowing incremental learning in an interactive environment. Artificial neural nets drive the classification of homogeneous text blocks. We have observed that when users use PLANET for ground truthing, the number of interactive corrections is quite large. In order to reduce user intervention and to make use of PLANET as a ground truthing system we have adapted its architecture.
Index Terms:
Ground Truth, Physical Layout Extraction, Datasets, Document Image, Artificial Neural Networks, Arabic Newspapers
Citation:
Karim Hadjar, Rolf Ingold, "Minimizing User Annotations in the Generation of Layout Ground-Truthed Data," icdar, pp.703-707, 2011 International Conference on Document Analysis and Recognition, 2011
Usage of this product signifies your acceptance of the Terms of Use.