[09-17]eu data protection laws-video KYC solution providers

A finger-vein imaging and liveness detection for identity authentication using 2-axis MEMS scanner

Date:2021-09-17

trulioo revenue

Author:client due diligence    kyc team

Keywords:aml onboarding,    using biometrics for security,    onfido agreement,    aml transaction monitoring systems,    money laundering compliance officer,  perform kyc

Description:

BT-5-2 Finger Liveness Detection Based on Pressure Dependence of Vein Image
Fingerprint identification systems have been widel
When the face recognition system is popularizedd,it is vulnerable to deception and attacks from non-real faces. The traditional methods for face liveness detection have drawbacks of long working time of network and high over-fitting possibility. Therefore,this paper proposes a new method based on convolutional neural network. The standard convolution kernel is replaced with a heterogeneous convolution kernel to detectreal faces. At the same time,in order to further optimize the classification effect,the functions of softmax loss and center loss are used as new classifiers. In the two classic databases of NUAA and CASIA-FASD,the proposed algorithm can improve accuracy and reduce training cost of the model. This algorithm shows higher efficiency than does the traditional convolutional neural network method.

Similar Articles