Abstract

This was my senior design project to complete my Bachelor’s at UNC Charlotte, completed under the advising of Dr. Hamed Tabkhi as a group with Kevin Chang, James Tallett, and Landon Gibson.

State-of-the-art deep learning models for computer vision have a wide range of applications in the public sector, including monitoring the usage of spaces for logistics and public health. However, “smart cities” which leverage artificial intelligence at a regional scale have considerations beyond improvements in model accuracy, such as privacy, security, and scalability. For our project, we worked to design and develop a prototype of a scalable and flexible AI system that addressed the needs of smart cities. The resulting system ran deep learning models on camera feeds in real-time, anonymized and stored data on AWS, and served end-users via a mobile app that could intuitively display population metrics at-a-glance.

Technologies Used

• Python
• Deep learning, computer vision
• Amazon Web Services, NoSQL databases
• Flutter

Under Construction!