Finger Vein Biometric Identification using Convolutional Neural Network and Electromyography
MS Antony Vigil1, Prashant Kumar2, Preetam Sarmah3, Rushab Kumar Jha4, Prashant Baheti5
1MS Antony Vigil, Department of Computer Science, SRM Institute of Science and Technology , Chennai (Tamil Nadu), India.
2Prashant Kumar, Department of Computer Science, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
3Preetam Sarmah, Department of Computer Science, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
4Rushab Kumar Jha, Department of Computer Science, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
5Prashant Baheti, Department of Computer Science, SRM Institute of Science and Technology, Chennai (Tamil Nadu), India.
Manuscript received on 18 April 2019 | Revised Manuscript received on 25 April 2019 | Manuscript published on 30 April 2019 | PP: 1493-1498 | Volume-8 Issue-4, April 2019 | Retrieval Number: D6536048419/19©BEIESP
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: A convolutional neural system (CNN) for finger-vein biometric verification is utilized here. The purpose behind utilizing this strategy is, not normal for existing biometric strategies, for example, unique finger impression and face, the vein designs are inside the body. Along these lines, making them for all intents and purposes difficult to duplicate or recreate. This makes finger-vein biometrics an increasingly secure option without being a hazard to falsification, harm, or change with time. In ordinary finger-vein acknowledgment strategies, different complex techniques to process picture to upgrade the picture is utilized to accomplish superior precision. In such manner, a critical preferred standpoint of the CNN over customary methodologies is its capacity to at the same time extricate highlights, diminish information dimensionality, and arrange in one system structure. What’s more, the strategy needs exclusively littlest picture preprocessing since the CNN is solid to commotion and modest misalignments of the non-heritable pictures. This assistance to keep the framework secure keep up the secrecy and trustworthiness of the framework without representing a hazard to the security of the framework.
Keywords: Convolutional Neural Network, Finger vein Biometrics, Biometric Authentication
Scope of the Article: Biomechanics