Euclidean and
Geodesic Distance between a Facial Feature Points in Two-Dimensional Face Recognition
System
Rachid Ahdid1,2,
Said Safi1, and Bouzid Manaut2
1Department of Mathematics and Informatics, Sultan
Moulay Slimane University, Morocco
2Poladisciplinary Faculty,
Sultan Moulay Slimane University, Morocco
Abstract: In
this paper, we present two features extraction methods for two-dimensional face
recognition. We have used the facial feature point detection to compute the
Euclidean Distance (ED) between all pairs of these points for the first
approach of Face
Feature Points (ED-FFP) and Geodesic
Distance (GD-FFP) in the second one. For a suitable comparison, we have
employed three well-known classification techniques: Neural Networks (NN),
k-Nearest Neighbor (KNN) and Support Vector Machines (SVM). To test the present
methods and evaluate its performance, a series of experiments were performed on
two-dimensional face image databases (ORL and Yale). Our results reveal that
the extraction of image features is computationally more efficient using GD
than ED.
Keywords: Face recognition, landmarks, ED, GD, neural networks, k-nearest neighbor and support vector machines.
Received February 22, 2017; accepted May 11, 2017