ORNAMENTAL LETTERS IMAGE CLASSIFICATION USING LOCAL
DISSIMILARITY MAPS
J. Landre´
Univ. of Reims-Champagne-Ardenne - CReSTIC
IUT, 9 rue de Que´bec, BP 396
10 026 Troyes cedex, France
jerome.landre@univ-reims.fr
F. Morain-Nicolier
Univ. of Reims-Champagne-Ardenne - CReSTIC
IUT, 9 rue de Que´bec, BP 396
10 026 Troyes cedex, France
frederic.nicolier@univ-reims.fr
S. Ruan Univ. of Reims-Champagne-Ardenne - CReSTIC
IUT, 9 rue de Que´bec, BP 396
10 026 Troyes cedex, France
su.ruan@univ-reims.fr
Abstract
This article describes a new method for ancient books
ornamental letters segmentation and recognition. The pur-
pose of our work is to automatically determine the letter
represented in an ornamental letter image. Our process is
divided in two parts: a segmentation step of the ornamental
letter is followed by a recognition step. The segmentation
process uses multiresolution analysis to filter background
decorations followed by a binarisation step and a morpho-
logic reconstruction of the expected letter. The recogni-
tion process use the previously obtained reconstruction and
compares it with capital letters images used as a dictionary
of shapes with the Local Dissimilarity Map (LDM) distance.
1 Introduction
Nowadays, cultural heritage preservation [8] is a very
important task for every nation in order to save knowledge
and literature for future generations. Ancient books from
the hand-press period often have beautiful page layouts be-
cause at this time, books were precious and rare objects.
Ornamental letters (or decorated initials) were used to start
new chapters to illustrate the content of the chapter. On
the web, many databases of ornamental letter images are
available for online search [5]. Efficient tools are needed to
browse these databases.
Figure 1 shows several ornamental letters images out of
the BVH [6] database.
Figure 1. Examples of ornamental letters
from the BVH database.
Our article is divided in seven parts. Section 2 describes
related work in the field. The local dissimilarity map tool
is presented in section 3. In section 4, a description of our
method is proposed. Detailed information on our technique
is developed in section 5. Experimental results are proposed
in section 6. At last, section 7 gives a conclusion and future
directions concerning our work.
2009 10th International Conference on Document Analysis and Recognition
978-0-7695-3725-2/09 $25.00 © 2009 IEEE
DOI 10.1109/ICDAR.2009.226
186