DocumentCode
3019830
Title
A synthetic database to assess segmentation algorithms
Author
Oliveira, Luiz S. ; Britto, Alceu S., Jr. ; Sabourin, Robert
Author_Institution
Pontificia Univ. Catolica de Parana, Curitiba, Brazil
fYear
2005
fDate
29 Aug.-1 Sept. 2005
Firstpage
207
Abstract
In this paper we describe a synthetic database composed of 273,452 handwritten touching digits pairs to assess segmentation algorithms. It contains several different kinds of touching and it was generated by connecting 2,000 images of isolated digits extracted from the NIST SD19. In order to get a better insight on the proposed database and establish some parameters for further comparisons, we carried out experiments using four state-of-the-art segmentation algorithms.
Keywords
handwriting recognition; handwritten character recognition; image segmentation; visual databases; handwriting recognition; handwritten touching digits pairs; state-of-the-art segmentation algorithm; synthetic database; Character recognition; Data mining; Handwriting recognition; Image databases; Image recognition; Image segmentation; Isolation technology; Joining processes; NIST; Robustness; Handwriting Recognition; Segmentation; Synthetic Data.;
fLanguage
English
Publisher
ieee
Conference_Titel
Document Analysis and Recognition, 2005. Proceedings. Eighth International Conference on
ISSN
1520-5263
Print_ISBN
0-7695-2420-6
Type
conf
DOI
10.1109/ICDAR.2005.35
Filename
1575539
Link To Document