Face Recognition System: Its Accuracy Can Be Counted Upon

By Author: Swaty Shah
Total Articles: 20
Comment this article

Advancement in technology has lead to improved security measures in firms, offices, companies, and other premises. Today, many prefer to use modern security equipment and devices to safeguard their premise. Surveillance cameras, fingerprint readers, and face recognition system are to name a few! The use of biometric face recognition has increased considerably in modern times. It is the most sought after technology in biometrics. It guarantees enhanced security of the premise.

The application makes use of complex algorithm for identity verification. Unique characteristics of the face are taken into consideration that does not age with time. There are eighty such nodal points on the face. Few of these points are considered by a quality face recognition system. The facial details of the enrolled individuals are stored in the database of the system as a face template. Next time onwards, when the individual looks at the camera, his facial details will be matched against the templates in the database. The enrolled individual will be recognized.

No security is foolproof. Biometric face recognition is no exception! ...
... It does not guarantee 100 percent accuracy. There have been certain controversies surrounding its excessive use. Many complain that it does not provide 100 percent shield. It is prone to errors and may commit blunders in recognizing enrolled individuals. This is true to some extent. Two recognized mistakes that this system commits can be as given below:

*False Accept Rate (FAR): This is considered to be a fatal error in the face recognition system. It gives unauthorized access to applications that may prove disastrous. Herein, the database contains templates of enrolled individuals. These are the only individuals meant to be recognized. However, the system may falsely identify those who are not enrolled as well. This gives them access to confidential data or premise, thereby, threatening the security of the individual.

*False Reject Rate (FRR): This case is bound to happen more than FAR. Herein, an individual might be enrolled in the face recognition system. However, due to some reasons, he may still be not recognized by the application. The reasons can be many; for example the lighting conditions may not be proper or the enrolled individual may not pose properly before the camera. It is not as disastrous but can be extremely frustrating to the user.

FAR and FRR reduces the accuracy in biometric face recognition technology. The type of reader used has a lot of impact in the overall accuracy. If the reader requires too much of facial details to be matched, there is a higher risk of FRR. If the reader requires too little of facial details to be matched, there is a higher risk of FAR. However, modern devices are designed with the possibility of all such errors in mind. Most of the vendors promise more than 90 percent accuracy in its use.

Modern applications cannot be fooled with a photograph either. The face recognition system only identifies live faces. It cannot be easily fooled or tricked. Though not completely infallible, it is more accurate than any other means of identity verification. In extremely sensitive places of importance, one cannot rely on the primitive modes of security. This application ought to be resorted to for the overall safeguarding of the area.

Author is related with face recognition system site justlook.co.in. Author often writes articles on Time Attendance System and Visitor Management System.