Facial Recognition with Open Cv

Abstract

In this digital era, the wide-range implementation of digital authentication techniques has not only impacted the home security systems but also in difference facets of industries. The ever-growing facial recognition technology is now dominated by the digital authentication system with a steady increase in the sophistication of the real time image processing operations, which is more accurate and reliable. The system is more secured by implementing facial recognition as a primary security feature for observing a person in a 3D perspective. In recent times, digital authentication system aims to recognize individuals, payment confirmation, criminal identification etc. In this research work, we have worked with a type of facial recognition system, which is based on “Open-CV” model. Open-CV is a library that is built by intel in 1999. It is a cross platform library, primarily built to work on real time image processing systems, which includes the state-of-the-art computer vision algorithms. The system that we have built can achieve an accuracy of about 85–90% accuracy based on the lighting condition and camera resolution. Based on the database, this system is built by taking still images of the subject, then it trains a model and creates deep neural networks and then for detection it matches the subject in real time and give results based on the match. The system that we have built is quite simple, robust and easy to implement, where it can be used in multiple places for a vast application.