GFRecog: A Generic Framework with Significant Feature Selection Approach for Face Recognition
Riyadh Rahef Nuiaa, Seif Ali Abdulhussein, Bahaa Kareem Mohammed · International Journal of Engineering & Technology · 2018
Identification of Humans uniquely is given paramount importance in the contemporary world. It is evident in applications of all fields so as to ensure secure and accurate transactions. Out of many approaches biometric approach became a dependable mechanism for this purpose. Face is one of the biometrics that plays vital role in recognizing humans across the globe. Many approaches came into existence for face recognition. In this paper we proposed a generic framework known as GFRecog that is extendable to support future methods of face recognition as well. We propose a methodology for face recognition using Gabor wavelets by extracting significant features from training dataset and perform matching operation with the given input image. Projection of face images onto a feature space that reflects diversity of face images is considered an efficient approach. Our approach works with faces that are captured under different lighting conditions, expression and pose. We built a prototype application using MATLAB with a benchmark dataset to demonstrate the proof of concept. The empirical results revealed that the accuracy of the proposed face recognition method is significantly high.