Scale-Invariant Face Recognition Using Triangular Geometrical Model

This work proposes a geometrical model based on multiple triangular features for the purpose of handling the challenge of scale variations that affect the process of face recognition especially in real time applications where the test images are usually taken in random scales that are usually different from the probe image. Geometrical approaches have proved to be robust to lighting and illumination variation. Furthermore geometrical methods in general do not hold computational complexity and have the benefit of faster processing time, which make them appropriate for real time applications. Fifteen triangle similarity measurement equations were derived and used to build a class of feature vectors for each subject. Ten images in ten different scales were taken for each subject for a total of fifty samples. Classification results show that the proposed model is promising in handling the challenge of scale variations.