Design and Implementation of Deep Learning Based Contactless Authentication System Using Hand Gestures

Hand gestures based sign language digits have several contactless applications. Applications include communication for impaired people, such as elderly and disabled health-care applications, automotive user interfaces, security surveillance. This work presents the design implementation of a complete end-to-end deep learning edge computing system that can verify contactlessly using ‘authentication code’. The code’ is an ‘n’ digit numeric code are hand digits. We propose memory-efficient model to classify proposed on bottleneck module which inspired by residual networks. achieves classification accuracy 99.1% publicly available dataset. deployed Raspberry pi 4 Model B serve device verification. consists two steps, it first takes input from camera attached in real-time stores buffer. In second step, classifies with inference rate 280 ms, taking image buffer input.