Image processing and face detection analysis on face verification based on the age stages

Abstract

The availability of computer systems has created a variety of automated applications in personal identification. From the various characteristics of biometrics, face recognition techniques mainly face verification has become an area of active research and the application are important in law enforcement because it can be done without involving the subject. However, the influence of age progression on face verification become a challenge to determine the similarity of image pairs from individual faces considering very limited of data base availability. This paper focuses on the development of image processing and face detection on face verification system by improving the image quality. The research use computer simulation, comparative studies, and analytical studies. The simulation of image processing using PhotoScape v3.5 software, while the verification using VeriLook 5.0/MegaMatcher VeriLook 4.0 Algorithm Demo. In the early step, database of face images are grouped by the age stages of education, then the simulation process includes the process of image processing, the enrolment process, and the verification process. The simulation results show the combination of image processing with contrast and sharpen increase the enrolment process by 17.14% with maximum success rate reaches 100%, where the entire 70-images training enrolled successfully. In verification process, the results of the combination increase the success rate of matching faces by 13.57%, where 96 image pairs successfully matched from a total of 140.