The Impacts of Depth Camera and Multi-Sensors for Face and Fingerprint Recognition in 2D and 3D - A Case Study

Abstract

This paper reviews the multi-modern sensors for multimodal (Face ad Fingerprint) of a biometric system by using a depth camera. The use of face authentication in biometric data allows this innovation to expand and be used in a variety of fields. Recently, attendance monitoring systems depending on biometric identification for higher education are underutilization, presenting an excellent chance to do fascinating experiments. The installation of biometric attendance systems necessitates the use of both hardware and software components. The first deep CNN (Convolutional Neural Network) method for light source-oriented Face Recognition (FR) takes advantage of the more detailed data given in a lens let display technology picture and that has been used in different borders such as airports, seaports, and land ports. In addition, the use of 3D camera technology for measuring medical outcomes in the healthcare marketplace is growing. The Intel® RealSense TM is one of the top 3D thermal imaging cameras systems on the market today, and it is well-suited for usage in a variety of areas such as medical systems, automation, and medical. Advances in in-depth sensor cameras technology have led to a considerable rise in the incorporation of these innovations into moveable systems, implying the great future potential for widespread in clinic and sector medical screening systems. Furthermore, initially, the use of dispersion maps in conjunction with depth maps and Two-Dimensional Red Green Blue (2D-RGB) pictures has been examined in terms of a fusion strategy to increase FR performance. The suggested approach employs the 2D-RGB crucial radiations emitted viewpoint, and depth maps and dispersion derived from the whole collection of radiations emitted pictures connected with a lens let light source. Following that, feature separation was carried out with the use of a VGG-deep sighs description for texturing and individually well their representations for depth maps and dispersion. The collected characteristics are then combined and supplied into a classification model.