Deep Learning, Face recognition, Spoofing detection, Convolutional Neural Networks.
Fingerprint and Face recognition systems are widely used in various applications like Smartphone unlock, School attendance, Defense applications, Banks, etc. However these systems can be spoofed easily using various methods like fake fingerprints can be created using various materials like Fevicol, Silicon gel, Hot glue, rubber, paper-printed fingerprints, etc and fake face can be created using paper-printed faces or a smartphone device can be used as a fake image. While a number of face and fingerprint spoof detection techniques have been proposed, current solutions often rely on domain knowledge, specific biometric reading systems, and attack types. This paper proposes using convolutional neural networks for detecting a spoofed fingerprint or face.
Since, CNNs currently outperform almost all other models for image recognition and classification. This paper proposes to use Google’s Inception v3 model which was trained on a huge dataset (ImageNet dataset) and has better acuracy than most other models.