An Iris Recognition System Using CNN & VGG16 Technique
Arun Singh, Akansha Pandey, Manik Rakhra, Dalwinder Singh, Gurasis Singh, Omdev Dahiya
Abstract
The human iris is a magnificent asset that can be used reliably for identifying purposes. It can eventually recognize humans with a serve degree of assertiveness. The extraction of enormous highlights is an essential component of the iris popularity framework. Previously, a variety of factors were used to run the iris popularity framework. The application of the capabilities acquired via the use of convolutional neural networks (CNNs) to iris recognition has attracted substantial interest due to the accomplishment of a high level of expertise in iris recognition. In this article, we investigate the capabilities of a convolutional neural network observed using the VGG16 method, often known as a convolutional community model. The entire performance of the advising device is evaluated with the extraction of capabilities from segmented and normalised iris images. The proposed iris popularity device is analysed using the CASIA-1000 dataset. The device provides incredibly effective effects at an exceedingly high rate of efficiency. On well-known iris datasets, the suggested method has been assessed and shown to achieve an accuracy rate of 96%, which surpasses the previous result.