A Three Layered Decentralized IoT Biometric Architecture for City Lockdown During COVID-19 Outbreak

Area of Science:

• Computer Science
• Artificial Intelligence
• Network Security

Background:

• Cities worldwide implemented lockdowns during COVID-19 to control virus spread.
• Biometric face detection is crucial for monitoring and enforcing movement restrictions.
• Existing solutions often struggle with scalability and latency in large-scale, real-time applications.

Purpose of the Study:

• To develop a decentralized Internet of Things (IoT) based biometric face detection framework.
• To implement a three-layered edge computing architecture for efficient face detection.
• To evaluate the proposed framework's performance against state-of-the-art methods and cloud computing architectures.

Main Methods:

• A deep learning framework utilizing multi-task cascading was developed for face recognition.
• The system employs a three-layered edge computing architecture for decentralized processing.
• Performance was benchmarked against FDDB and WIDER FACE datasets.
• Latency and face detection load experiments were conducted comparing edge and cloud computing.

Main Results:

• The proposed decentralized IoT framework demonstrated effective biometric face detection.
• The three-layered edge computing architecture showed significant advantages over traditional cloud computing.
• Experiments confirmed lower latency and better load handling in the edge computing model.

Conclusions:

• The developed framework provides an efficient and scalable solution for biometric face detection during city lockdowns.
• Edge computing offers a superior alternative to cloud computing for real-time public movement monitoring in critical situations.
• This research contributes to the advancement of smart city technologies for public health crisis management.