Iris identification, how is it done?

The last biometric that we will use is the output of an iris scan. Considering our setup, there might be several ways to grab iris data:

• We can separate the face and apply an eye detector using face detection, which can be done with a high-resolution camera. We can use the resulting regions to perform iris segmentation and classification.

• We can use a specific eye camera, which grabs an eye image to be classified. This can be done either with RGB or NIR.

Since the first approach is prone to a lot of problems, such as the resulting eye image having a low resolution, a more common approach is to use a separate eye camera that grabs the eye. This is the method that we will use in this chapter. An example of a captured eye in both the RGB (visible colors) and NIR (near infra-red) spectrums is visualized below:

An example of both a RGB and a NIR iris-based image

Using NIR images helps us in several ways:

• First of all, color information is omitted, since a lot of...