Geodesic
Classification and Clustering of Printed Mathematical Symbols with Improved Backpropagation andSelf-Organizing Map
Bulletin of the Malaysian Mathematical Society, Tome 22 (1999) no. 2 
Siti Mariyam Hj. Shamsuddin; Md. Nasir Sulaiman; Maslina Darus. Classification
                          and Clustering of Printed Mathematical Symbols with
                          Improved Backpropagation andSelf-Organizing Map. Bulletin of the Malaysian Mathematical Society, Tome 22 (1999) no. 2. http://geodesic.mathdoc.fr/item/BMMS_1999_22_2_a5/
@article{BMMS_1999_22_2_a5,
     author = {Siti Mariyam Hj. Shamsuddin and Md. Nasir Sulaiman and Maslina Darus},
     title = {Classification
                          and {Clustering} of {Printed} {Mathematical} {Symbols} with
                          {Improved} {Backpropagation} {andSelf-Organizing} {Map}},
     journal = {Bulletin of the Malaysian Mathematical Society},
     year = {1999},
     volume = {22},
     number = {2},
     url = {http://geodesic.mathdoc.fr/item/BMMS_1999_22_2_a5/}
}
TY  - JOUR
AU  - Siti Mariyam Hj. Shamsuddin
AU  - Md. Nasir Sulaiman
AU  - Maslina Darus
TI  - Classification
                          and Clustering of Printed Mathematical Symbols with
                          Improved Backpropagation andSelf-Organizing Map
JO  - Bulletin of the Malaysian Mathematical Society
PY  - 1999
VL  - 22
IS  - 2
UR  - http://geodesic.mathdoc.fr/item/BMMS_1999_22_2_a5/
ID  - BMMS_1999_22_2_a5
ER  - 
%0 Journal Article
%A Siti Mariyam Hj. Shamsuddin
%A Md. Nasir Sulaiman
%A Maslina Darus
%T Classification
                          and Clustering of Printed Mathematical Symbols with
                          Improved Backpropagation andSelf-Organizing Map
%J Bulletin of the Malaysian Mathematical Society
%D 1999
%V 22
%N 2
%U http://geodesic.mathdoc.fr/item/BMMS_1999_22_2_a5/
%F BMMS_1999_22_2_a5

Voir la notice de l'article provenant de la source BMMS

This paper proposes a derivation of an improved error signal for hidden layer in the backpropagation model, and its experimentation evaluation of utilizing various moments order as pattern features in recognition of printed mathematical symbols in the classification phase. The moments that have been used are geometric moment invariants in which they have been used as feature extraction for images with various orientations and scaling. In this study, we find that the recognition and the convergence rates are better using an improved backpropagation compared to standard backpropagation. In addition, we cluster these invariants on a visual map using Self-Organizing Map (SOM) whereby mathematical symbols with similar shape belong to the same cluster.