Communications of the ACM
Artificial Intelligence and Supercomputers to Help Alleviate rban Traffic Problems
Researchers at the Texas Advanced Computing Center (TACC), the University of Texas Center for Transportation Research, and the city of Austin are using deep learning and data mining to develop tools enabling searchable traffic analyses.
Credit: Texas Advanced Computing Center/Center for Transportation Research
Researchers at the Texas Advanced Computing Center (TACC), the University of Texas Center for Transportation Research, and the city of Austin are working to develop tools enabling searchable traffic analyses using deep learning and data mining.
One tool uses raw traffic camera footage to recognize objects such as people, cars, bicycles, and traffic lights and characterize how those objects move and interact. The data can be analyzed and queried by traffic engineers and officials to determine, for example, how many cars drive the wrong way down a one-way street.
"We want to explore means that may be helpful for a number of analytical needs, even those that may pop up in the future," says TACC's Weijia Xu. He notes the traffic analysis algorithm automatically labels all potential objects from the raw data, tracks objects by comparing them with other previously recognized objects, and measures the outputs from each frame against the others to determine the relationships among the objects.
From UT News
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found