Committee Chair

Sartipi, Mina

Committee Member

Liang, Yu; Kandah, Farah

Department

Dept. of Computer Science and Engineering

College

College of Engineering and Computer Science

Publisher

University of Tennessee at Chattanooga

Place of Publication

Chattanooga (Tenn.)

Abstract

With the predicted boom of urban environment populations in the next 30 years, many new challenges in urban transportation will surface. In an effort to mitigate these, the Center for Urban Informatics and Progress (CUIP) has been introduced along with its testbed. One opportunity this testbed provides is the ability to utilize computer vision and video analytics to anonymously gather data on how citizens traverse the city. This thesis shall discuss an approach to real-time object tracking that serves as a basis for further analytics such as traffic flow data collection and near-miss detection. The proposed video analytics platform will aid citizens with their day-to-day commute through the corridor by deriving real-time data based on actual behavior seen in the citizens' commute. Furthermore, since the testbed is ever-expanding in both hardware and size the algorithms and software proposed in this thesis are designed to prioritize scalability.

Acknowledgments

I would like to begin with thanking my family (all of them, including those that were chosen). Your support has motivated me to become the best individual that I can be, and without them I would be lost. I would like to extend my gratitude to Dr. Farah Kandah for his guidance and support throughout my graduate-level education. Additionally, I would like to thank Dr. Yu Liang for giving me support and motivation to continue my work and understand the underlying theory in otherwise black-boxed software implementations. I would like to thank the team members of the lab when I originally joined in 2017, all of which helped me learn new concepts quicker than I ever have! Specifically, I would like to thank Dr. Zhen Hu, Rebekah Thompson, Jin Cho, Austin Harris, and Hector Suarez. Together, they helped keep me learning the many concepts I was so unfamiliar with at the time and supported me the entire time. I would also like to thank the current lab team for their support (both emotional and otherwise). These exceptional individuals have been there to help me understand concepts I was stuck on, give me a good laugh when I needed it, and help me up when I was down. Specifically, I’d like to thank Dr. Thanh Nam Doan, Peter Way, Jeremy Roland, Katie Rouse, Bennett Bowden, Yatri Patel, Alnour Alharin, and Sree Nukala. I cannot extend enough thanks to thank Dr. Mina Sartipi, without whom I would have never come as far as I have. Through her, I have built confidence in my work and ability to develop cutting-edge algorithms that I would have never dreamt of doing otherwise. I am forever grateful for the opportunity she has given me, first at SCAL then at CUIP, and I cannot thank her enough for her major contribution to who I have become, both as an individual and as a professional.

Degree

M. S.; A thesis submitted to the faculty of the University of Tennessee at Chattanooga in partial fulfillment of the requirements of the degree of Master of Science.

Date

5-2020

Subject

Computer vision; Traffic conflicts; Traffic flow

Keyword

Video Analytics; Data Science; Computer Vision; Object Tracking; Near-Miss; Traffic Flow

Document Type

Masters theses

DCMI Type

Text

Extent

xi, 58 leaves.

Language

English

Rights

http://rightsstatements.org/vocab/InC/1.0/

License

http://creativecommons.org/licenses/by/4.0/

Download

Share

COinS