An assessment of face recognition system for border security / Mohamed Ishan

Abstract

In this modern world, thousands of people travel from one country to another country every day. Travelling has become one of the basic needs for many people for many different reasons. And within these people there are many who had been in the wanted list for different criminal activities which might had been done in different parts of the world.

For this reason, one of the most secured and monitored areas in all the countries is the immigration security. It is very important to securely monitor people entering to the country. Most of the countries, people entering the country are identified using their passport. Using this information immigration personal can identify blacklisted personals if their passport had been blacklisted. But what if the only available information of the criminal is only an image of that person?

Biometrics system has been around for a long time and one of the widely used in the market is thumb
verification systems. Thumb verification is a very reliable and effective way of uniquely identifying people. But if the fingerprint information is not available then this method is not very effective. The only effective method to solve this problem is to identify the person using face recognition.
In most cases border security officials will be provided with a list of photos of the criminals. As humans this makes it very difficult to identify such criminals. To identify such criminals, border security officials has to memorize and remember all the images of the criminals provided to them. Matching all the travelers with these list of images is a very un-effective way and the failure rate of identifying the criminals this way is very high. For this reason, another alternative is needed to overcome this problem

Keywords: Face Recognition System, Eigenface, Image Processing, Mean Filtering, Median Filtering