System Designing For Face Spoofing Detection Using Convolution Neural Network (Cnn)

Authors

  • Wafaa Ali Abdulhussein Alrikabi , Ahmed Muqdad Alnasrallah, Hanan Ali Alrikabi

Abstract

the face image is that the most accessible biometric modality that is employed for immensely precise face recognition systems, whereas it's exposed to many various types of staging attacks. Face anti-spoofing may be a terribly essential step before feeding the face image to biometric systems. During this research, we tend to gift a completely unique CNN-based approach for face anti-spoofing, by extracting the native options from the face images. The local features facilitate CNN to discriminate the spoof patches freelance of the spatial face areas. We create our custom dataset for this purpose and the result demonstrated that the proposed system work properly in terms of detecting spoofing attack on printed picture or mobile video etc. the testing accuracy of the proposed research work using custom data set is 96% which is highly appreciable.

Published

2020-12-01

Issue

Section

Articles