Performance Evaluation of Different Feature Extractors and Classifiers for Recognition of Human Faces with Low Resolution Images
Journal Title: International Journal of Intelligent Systems and Applications in Engineering - Year 2015, Vol 3, Issue 2
Abstract
Face recognition is an effective biometric identification technique used in many applications such as law enforcement, document validation and video surveillance. In this paper the effect of low resolution images which are captured in real world applications, on the performance of different feature extraction techniques combined with a variety of classification approaches is evaluated. Gabor features and its combination with local phase quantization histogram (GLPQH) are dimensionality reduced by principal component analysis (PCA), linear discriminant analysis (LDA), locally sensitive discriminant analysis (LSDA) and neighbourhood preserving embedding (NPE) to extract discriminant image characteristics and the class label is attributed using the extreme learning machine (ELM), sparse classifier (SC), fuzzy nearest neighbour (FNN) or regularized discriminant classifier (RDC). ORL and AR databases are utilized and the results show that ELM and RDC have better performance and stability against resolution reduction, especially on Gabor-PCA and Gabor-LDA techniques. Among the interpolation approaches that we employed to enhance the image resolution, nearest neighbour outperforms other methods.
Authors and Affiliations
Soodeh Nikan*| ECE, University of Windsor, Windsor, ON – N9B 3P4, Canada, Majid Ahmadi| ECE, University of Windsor, Windsor, ON – N9B 3P4, Canada
Keywords
Related Articles
- EP ID EP773
- DOI 10.18201/ijisae.28949
- Views 479
- Downloads 22
A fuzzy approach for determination of prostate cancer
Goal of this study is a design of a fuzzy expert system, its application aspects in the medicine area and its introduction for calculation of numeric value of prostate cancer risk. For this aim it was used prostate speci...
Lossless Text Compression Technique with LSB Technique to Hide Secret Message inside an Image (CLSB)
This paper presents CLSB algorithm to improve and increase the security ofhiding message inside an image by using Least Significant Bit (LSB) method. This research attempts to improve the way has been introduced in [1],...
Grade prediction improved by regular and maximal association rules
In this paper we propose a method of predicting student scholar performance using the power of regular and maximal association rules. Due to the large number of generated rules, traditional data mining algorithms can bec...
Fusion of Target Detection Algorithms in Hyperspectral Images
Target detection in hyperspectral images is important in many applications including search and rescue operations, defence systems, mineral exploration and border security. For this purpose, several target detection algo...
The Classification of Diseased Trees by Using kNN and MLP Classification Models According to the Satellite Imagery
In this study, the Japanese Oak and Pine Wilt in forested areas of Japan was classified into two group as diseased trees and all other land cover area according to the 6 attributes in the spectral data set of the forest....