Video Analytics Traffic Study Creates Baseline for Change
Bellevue, Wash., conducted a traffic study examining thousands of hours of video footage taken from 40 intersections for data around near-miss accidents. The project is part of a larger effort to reduce traffic fatalities.
/ JULY 21, 2020
Cyclists in one Washington city are 10 times more likely to be involved in a traffic conflict than motorists at a number of key intersections.
This finding and others were part of a recent report of near-miss crashes in Bellevue, Wash., examining video data collected from high-definition traffic cameras. The study, known officially as the Video-based Network-wide Conflict and Speed Analysis to Support Vision Zero in Bellevue (WA) United States, also found that 10 percent of motorists are speeding and motorcycles “traveled at higher speeds and generated more critical conflicts than any other road user.”
The study analyzed some 4,500 hours of video footage across 40 intersections, taking in more than 8 million road users, and observing some 20,000 near-misses. The city adopted its Vision Zero policy in 2015, and video analytics is part of that effort, say officials.
“By measuring vulnerable road user safety at-scale, in ways that weren’t possible even a few years ago, we’re making the invisible visible and giving city leaders foundations to make data-driven decisions around infrastructure and safety,” said Noah Budnick, senior director for programs and operations at Together for Safer Roads, a partner on the study, along with Transoft Solutions Inc.
Video data in the study, collected from mid-September of last year, helped to generate three analyses which offered added insight into near-miss incidents, say transportation officials. The study will help guide policy proposals related to biking and pedestrian infrastructure, as well as signage and other street improvements, said David Grant, public information officer for the Bellevue Transportation Department.
“I think the main policy discussion will come when budget proposals are considered later this year. There are specific proposals for new pedestrian and bicycle infrastructure,” Grant explained. “The study will help us identify and prioritize projects on the ground.”
Even though the data and analytics behind the street safety study were gathered in September 2019, it’s been presented in the midst of the COVID-19 pandemic, which has in many cities reshuffled and reshaped the streetscape, giving a previously unheard of amount of precedence to active transportation options in many cities by reducing car access on a number of streets.
Some of the changes like street closures and increased cycling may stick around well after COVID-19, and make micro-mobility a stronger option and continue to make cities more resilient, said Ron Burke, head of active transportation at Lyft, speaking on a panel during the Shared-Use Mobility Center virtual summit in early May.
“Truly resilient cities are not overly dependent on cars, aren’t overly dependent on personal car ownership,” said Burke, “and make it possible for people to get around with a variety of modes.”
“In addition to exercise and clean air, the pandemic has highlighted the urgent public health benefits of walking, bicycling and scooting — physical distance and open air, among others,” added Budnick, in an email. “We hope visual analytics like this shows leaders how they can collect data and use it to support people who choose these modes. Communities need mobility, but we also need to desperately mitigate the highly contagious nature of the coronavirus, and many are realizing that walking, biking and scooting do this.”
Never far from any discussion around the use of mobility data and the capture of video imagery in the public space are concerns about how this data is used and shared. The Transoft technology used by Bellevue’s traffic analytics is not capable of facial recognition or license plate monitoring, said Grant.
“Bellevue’s MOU [memorandum of understanding] with Transoft prohibits using video for these purposes,” he added.
The city uses red light cameras and school zone cameras for “photo enforcement.” And video made available as part of public records requests is reviewed and redacted before its release to shield personal information, said Grant.
Bellevue gathers up transportation data from a variety of sources: bicycle counts, police incident reports, resident surveys and other materials.
“But none of these can offer the same level of data as traffic analytics when it comes to measuring traffic volume, speeds and conflicts,” said Grant, adding, “The city does not use data from mobile phone movements.”
Findings from the Video-based Network-wide Conflict and Speed Analysis to Support Vision Zero in Bellevue study:
- 38 percent of crash causes can be explained by a combination of road user volume and conflict data.
- Motorcyclists travelled at higher speeds and generated more critical conflicts than any other road user.
- Cyclists made up 0.1 percent of observed road users, with motorists accounting for 97.3 percent and pedestrians 2.6 percent.
- More than 10 percent of drivers were speeding. And half of these drivers speeding were traveling in excess of 11 mph over the posted speed limit.
Source: City of Bellevue