Over a million developers have joined DZone.
{{announcement.body}}
{{announcement.title}}

Using AI to Build Smarter Traffic Models

DZone's Guide to

Using AI to Build Smarter Traffic Models

See how AI can be effectively used to provide more comprehensive and accurate analyses of traffic data to provide decision-makers with actionable insights.

· AI Zone ·
Free Resource

Did you know that 50- 80% of your enterprise business processes can be automated with AssistEdge?  Identify processes, deploy bots and scale effortlessly with AssistEdge.

City planners already receive significant amounts of data about traffic flows, but congestion is still a problem that blights many cities today. A team from the Texas Advanced Computing Center (TACC), the University of Texas Center for Transportation Research, and the City of Austin believes that AI can help to optimize traffic flow.

Their work is developing searchable traffic analyses using deep learning and data mining. The tool uses the raw data generated from traffic cameras, with their algorithm capable of recognizing the various objects in the footage and then characterizing how those objects move and interact. This information can then be queried by traffic planners to better understand the transport network.

"We are hoping to develop a flexible and efficient system to aid traffic researchers and decision-makers for dynamic, real-life analysis needs," the team says. "We don't want to build a turn-key solution for a single, specific problem. We want to explore means that may be helpful for a number of analytical needs, even those that may pop up in the future."

Smarter Traffic Management

The system is capable of automatically tagging each object it encounters in the raw data and then tracking their movements throughout the footage. By comparing outputs from each frame, the algorithm is able to discover relationships among the objects.

The system was put through its paces in two practical use cases. The first saw it count the number of vehicles traveling down a road while the second identified near-misses between vehicles and pedestrians. The results showed that the algorithm was around 95% accurate when counting the vehicles.

The system automatically counted vehicles in a ten-minute video clip, and preliminary results showed that their tool was 95% accurate overall. The team believes that the initial findings validate the system, which will be crucial (and cost-effective) in assessing traffic models in use today.

"Current practice often relies on the use of expensive sensors for continuous data collection or on traffic studies that sample traffic volumes for a few days during selected time periods," they explain. "The use of artificial intelligence to automatically generate traffic volumes from existing cameras would provide a much broader spatial and temporal coverage of the transportation network, facilitating the generation of valuable datasets to support innovative research and to understand the impact of traffic management and operation decisions."

Spotting Near-Misses

The system was also capable of accurately spotting a number of near-misses between vehicles and pedestrians. As none of these incidents were real-life dangers, they probably would go undetected in existing traffic monitoring systems, but they nonetheless help to identify dangerous locations and hopefully allow staff to intervene before injuries occur.

The team hopes to continue developing the platform so that it can autonomously analyze other safety-related things, such as where pedestrians cross the road unsafely.

It's a nice example of how AI can be effectively used to provide more comprehensive and accurate analyses of the kind of traffic data that most cities are generating today to provide decision-makers with actionable insights.

"The highly anticipated introduction of self-driving and connected cars may lead to significant changes in the behavior of vehicles and pedestrians and on the performance of roadways," the team concludes. "Video data will play a key role in understanding such changes, and artificial intelligence may be central to enabling comprehensive large-scale studies that truly capture the impact of the new technologies."

Check out the video below to see the system in action.

Consuming AI in byte sized applications is the best way to transform digitally. #BuiltOnAI, EdgeVerve’s business application, provides you with everything you need to plug & play AI into your enterprise.  Learn more.

Topics:
ai ,traffic ,data analytics ,deep learning ,data mining ,algorithm

Published at DZone with permission of

Opinions expressed by DZone contributors are their own.

{{ parent.title || parent.header.title}}

{{ parent.tldr }}

{{ parent.urlSource.name }}