Close this search box.

An Evolving Career In Computer Vision Leads Students into the Future of CAVs Technology

The University of Tennessee, Knoxville EcoCAR team has been working diligently towards re-engineering their 2019 Chevrolet Blazer during Year 3. The main focus for this year has been the integration of Connected and Automated Vehicle (CAVs) technology. The UTK CAVs subteam, overseen by CAVs Faculty Advisor Dr. Hairong Qi, has been integrating cameras and radar onto their vehicle. Dr. Qi has guided the team through these activities by utilizing her experience and research on the evolution of computer vision.

Dr. Qi is the director of the Advanced Imaging and Collaborative Information Processing (AICIP) Lab at the University of Tennessee, Knoxville. Her main area of research is computer vision and machine learning. Dr. Qi received her Bachelor of Science in Computer Science and Master of Science in Computer Science from Northern JiaoTong University, Beijing, China, in 1992 and 1995. Qi went on to receive her PhD in Computer Engineering from North Carolina State University, Raleigh, NC, in 1999.

Dr. Qi’s background has evolved from an interest with single image processing began over 20 years ago. Image processing, which applies mathematical functions to images, involves sharpening, stretching, smoothing or contrasting an image. Dr. Qi’s research consists of how to make an image more clear with less blur and noise. These primary activities led to an interest in systems with a wide network of sensors. Most recently, Dr. Qi’s research evolved from collaborative processing of 1-D signals to 2-D images and to x-D multimedia sources within a complex cyber-physical system (CPS).

Dr. Qi’s research has sense evolved to computer vision, which is when systems attempt to replicate human vision by utilizing computers. These systems assist computers with learning and taking actions from visual data. Computer vision utilizes artificial intelligence (AI) and machine learning to complete these actions.

You may ask, how does Dr. Qi’s interest and expertise relate to EcoCAR? Autonomous driving vehicles are an example of a complex cyber-physical system. Perception and fusion of data collected from the onboard sensor suite of a vehicle is one key component that enables autonomous driving. During autonomous driving, systems identify and sort various objects by using cameras and sensors embedded within a vehicle. Computer vision then identifies these images and videos and assigns them a specific size, shape or color. Computer vision can then produce results for a driver in real-time.

This interest in autonomous vehicle technology and its use of complex computer vision led Dr. Qi to the EcoCAR Mobility Challenge program at UTK. Specifically, Qi is drawn to the rise of deep learning which has enabled an unprecedented leap forward in computer vision, especially in object detection, recognition, and tracking, which are the key components in autonomous driving. She found interest in helping students apply these technologies into a vehicle for experimental learning purposes.

“Each day I am in awe by the amount of hands-on exposure students have working on the project – every component in the car can be removed and replaced to update the functionality of the car,” Qi said. Dr. Qi enjoys teaching the younger generation about computer vision and connected and automated vehicles. “To be able to follow the entire integration process of CAVs into the team vehicles and be part of it is a great way of learning,” she stated.

One of her favorite parts of the project is perhaps the amount of growth from students, both personally and professionally. “I’m fascinated by how much passion and independence the student team shows. You get a feel that this is THEIR project, their baby and you are working for them. I’m proud of our student team,” said Qi.

Although Dr. Qi’s research has evolved much over the last decade, she looks ahead to the future of autonomous vehicles with both caution and optimism. “I feel that researchers in computer vision can solve autonomous vehicles more effectively if a certain level of vehicle background can be gained,” Qi said. “Ultimately, no matter what algorithms we design, it serves the purpose of autonomous driving, be it better fuel efficiency, better mobility, better safety,” she added.

Dr. Qi has earned numerous awards for her research and 200+ scientific papers, she has authored two books, and has received recognition from the National Science Foundation (NSF) and the Institute of Electrical and Electronics Engineers (IEEE). Qi has been able to use her skills and decades of experience to teach the blossoming young engineering students about CAVs technology.

In addition to her EcoCAR involvement, Qi enjoys cooking, reading, and traveling, and on a rainy day you can catch her inside with a novel and a cup of tea.

Related Posts