Earprint Based Mobile User Authentication Using Convolutional Neural Network and SIFT
Biometric verification techniques are increasingly being used in mobile devices these days with the aim of keeping private data secure and impregnable. In our approach, we propose to use the inbuilt capacitive touchscreen of mobile devices as an image sensor to collect the image of ear (earprint) and use it as biometrics. The technique produces a precision of 0.8761 and recall of 0.596 on the acquired data. Since most of the touch screens are capacitive sensing, our proposed technique presents a reliable biometric solution for a vast number of mobile devices.
KeywordsMobile biometrics Earprint Convolutional neural network SIFT
- 3.Okumura, F., Kubota, A., Hatori, Y., Matsuo, K., Hashimoto, M., Koike, A.: A study on biometric authentication based on arm sweep action with acceleration sensor. In: Proceedings of International Symposium on Intelligent Signal Processing and Communication, pp. 219–222 (2006)Google Scholar
- 6.Holz, C., Buthpitiya, S., Knaust, M.: Bodyprint: biometric user identification on mobile devices using the capacitive touchscreen to scan body parts. In: Proceedings of Annual Conference on Human Factors in Computing systems, pp. 3011–3014 (2015)Google Scholar
- 7.Descartes Biometrics. http://www.descartesbiometrics.com/helix-sdk/
- 9.Li, Y., Huang, J.-B., Ahuja, N., and Yang, M.-H.: Joint image filtering with deep convolutional Networks ArXiv e-prints, arXiv:1710.04200 (2017)