Iris detection using morphology

Abstract EN

For accurate iris recognition, it is essential to detect iris.

Iris recognition systems are mainly dependent on the performance of iris localization processing.

So, the goal of this work is iris detection, where it is suggested using morphological operations (erosion and dilation) in iris detection.

Erosion is used to erode away the boundaries of undesirable regions of foreground pixels in annular of iris that resulted from threshold process.

Whereas, dilation is gradually used to enlarge the boundaries of regions of foreground pixels such as pupil who shrunk by erosion operation.

Then, smoothing the edges by using mean filter.

The significant drawbacks of this method are that some operations were manipulated manually such as; removing the eyelids and eyelashes from the original image.