Title : Gender identification using 3D eyeball pupil model
Abstract:
3D iris recognition is one of the most reliable techniques for identifying the gender of individuals. Each iris consists of an inner and outer circular looking area. The outer area does not change in size, while the inner area can change in size and appearance over time. These changes in the inner area make it difficult to use in biometric recognition and authentication systems. The centre of the iris is called the pupil, an opening through which light passes. The iris adjusts the size of the pupil to control the amount of light that enters the eye. In the system detailed here, the pupil is estimated as the main feature and calculated as a reference in the system to recognise the gender. It works on the pupil of the iris to measure the diameter of the targeted pupil and identifies the gender of the input image from the standard dataset. The proposed system produces accurate results and more usability in the system. The proposed system offers a simplistic technique and does simple formulas calculations compared to the classification using the Low Binary Pattern (LBP), texture patterns and five-layer deep convolutional neural network. The system has been tested using the Multimedia University Iris database and classified individuals in images as either male or female with an accuracy rate of 92%.
