Elyazisi Tanima Sistemlerinde Doǧrusal Boyut İndirgeme Yöntemleri

Translated title of the contribution: Linear dimension reduction methods in character recognition systems

Abdülkerim Çapar*, Alper Ayvaci, Fatih Kahraman, Hakan Demirel, Muhittin Gökmen

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Handwritten Character Recognition Systems can be divided into three steps: pre-processing, feature extraction and classification. In the feature extraction process, representation power of features should be increased yet keeping the number of features as small as possible. In this study, raw character image vectors are projected to lower dimension spaces by different linear transformations and their representation and discrimination power are compared. In dimension reduction, Principal Component Analysis (PCA), Multiple Discriminant Analysis (MDA) and Independent Component Analysis (ICA) are compared and best classification performance is obtained by using ICA. A multi layer perceptron, which is trained by conjugate gradient algorithm, is used for classification. The handwritten character database, studied on, consists of 5000 training patterns and 2500 test patterns.

Translated title of the contributionLinear dimension reduction methods in character recognition systems
Original languageTurkish
Title of host publicationProceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004
EditorsB. Gunsel
Pages611-614
Number of pages4
Publication statusPublished - 2004
EventProceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004 - Kusadasi, Turkey
Duration: 28 Apr 200430 Apr 2004

Publication series

NameProceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004

Conference

ConferenceProceedings of the IEEE 12th Signal Processing and Communications Applications Conference, SIU 2004
Country/TerritoryTurkey
CityKusadasi
Period28/04/0430/04/04

Fingerprint

Dive into the research topics of 'Linear dimension reduction methods in character recognition systems'. Together they form a unique fingerprint.

Cite this