Classification of face images using discrete cosine transform

In this study, it is aimed to determine whether a given image belongs to for that person. For feature extraction, which is an important part of pattern recognition, feature vector is obtained by using discrete cosine transform after performing preprocess the images on the current face. Based on the...

Full description

Saved in:
Bibliographic Details
Main Authors: Karhan, Z., Ergen, B.
Format: Conference Proceeding
Language:eng ; tur
Subjects:
Abstracts
Classification algorithms
Discrete cosine transforms
discrete cosinüs transform
Face
Feature extraction
K NN Classifier
Pattern recognition
Support vector machine classification
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In this study, it is aimed to determine whether a given image belongs to for that person. For feature extraction, which is an important part of pattern recognition, feature vector is obtained by using discrete cosine transform after performing preprocess the images on the current face. Based on the of datas obtained from conversion are classified by using 5%, 8%, 10%, and 15%. The nearest neighbor algorithm (KNN) is used in classification process. Face images consist of images that, taken from ORL database, belongs to 40 individuals, each has 10 different images. As a result, high success were obtained by using the few data.
DOI:10.1109/SIU.2013.6531364