Fireseeds North Infrastructure

UN Road Safety Week

Fireseeds North Infrastructure is offering 100 free video-based
Safe Systems speed studies
at intersections around the world to
mark the fourth United Nations Global Road Safety Week.

 
 

The Project


Speeds measured using automated video-analytics software

Speeds measured using automated video-analytics software

Fireseeds North Infrastructure (FNI) is offering 100 free speed studies at intersections around the world to mark the 2017 United Nations Global Road Safety Week. This page contains information on the project including how you can register to participate. Registration is open until June 30, 2017 and there is no cost to participate. Participants are responsible for collecting and transmitting video data to FNI at one or more signalized intersections of their choice.

At each intersection, FNI will use automated video-analytics software to determine speed profiles of left turning vehicles at downstream crosswalks. FNI’s video-analytics software automatically tracks objects, such as vehicles or pedestrians, and analyzes their trajectories through time and space to determine speeds and other safety indicators.  

Speed profiles will be compared with biomechanical injury thresholds for pedestrians to quantify risk associated with potential left turn pedestrian crashes.

Participants will receive a final report containing left turn speed profiles at their intersection(s), comparisons with biomechanical injury thresholds for pedestrians, and a list of evidence-based treatments for mitigating left turn pedestrian crashes.

FNI will also collectively analyze speed profiles across all intersections in a research project to better understand left turn driver behaviour and speeds. The results of this analysis will be shared with participants and submitted for publication in various journal and conference forums.

 

Background


The Relationship between Speed and Crash Severity

Road traffic crashes are a global health epidemic resulting in approximately 1.25 million fatalities and millions more injuries every year. This burden is disproportionately placed on vulnerable road users, like pedestrians, cyclists, and motorcyclists, who account for around half of all global road traffic fatalities. Speed is a key risk factor in crash likelihood and severity, especially for vulnerable road users, and is the focus of this year’s United Nations Global Road Safety Week which occurred May 8–14, 2017.

The Safe Systems Approach to road safety recognizes that the human body is highly vulnerable to injury with a known tolerance to withstand physical force, and that humans are fallible and make mistakes. By reducing vehicle speeds, we reduce road user error by providing more time for drivers to perceive and react to conditions. We also reduce the impact forces released in a crash which improves chances for survivability.

 

The higher the speed of the vehicle, the higher the risk of injury and death. Pedestrians face a 10% risk of death at 30 km/h vs. 80% risk of death at 50 km/h.

 
Infographic2.png
 
 

Left Turn Pedestrian Crashes

Pedestrian crashes with left turning vehicles at signalized intersections are one of the most common pedestrian crash types: left turns account for more than twice as many pedestrian fatalities as right turns, and more than three times as many serious injuries and fatalities combined (New York City Vision Zero, 2017).

Left turn pedestrian conflict

Left turn pedestrian conflict

Some factors which contribute to left turn pedestrian crashes at signalized intersections include:

  • The mental workload for drivers is greatest for left-turning movements (Lord, Smiley, & Haroun, 1998). Drivers must focus on understanding signal indications, judging and accepting gaps in oncoming traffic, and responding to downstream crossing pedestrians who may only be visible in their peripheries.
  • Large turn radii and longer clearance distance allows left turn drivers to achieve higher speeds, which increases the potential severity outcomes of a crash.
  • Driver visibility to pedestrians may be obscured by left turning vehicles ahead, parked vehicles, and the vehicle’s A-pillar, which is the portion of the vehicle which connects the vehicle frame to the windshield.
  • Signal operations may permit left turn movements concurrently with downstream pedestrian crossings, which creates the opportunity for conflict.
  • The design walking speed used to time the pedestrian clearance interval may not accommodate older pedestrians or pedestrians with mobility impairments, who travel slower. These pedestrians may not have sufficient time to cross and could be left within the intersection when conflicting motor-vehicle traffic is present.

An improved understanding of left turn driver behaviour and speeds at downstream crosswalks can help determine how to best mitigate these conflict types and reduce potential severity outcomes of pedestrian crashes.

 

Video Data Collection Specifications and Requirements


Participants are responsible for collecting and transmitting video data to FNI for analysis. Video data must meet the requirements indicated below to be successfully analyzed in the software. FNI will evaluate speed profiles of left turning vehicles at all downstream crosswalks that are sufficiently visible in the video: it may not be possible to evaluate speeds at all crosswalks.

Participants may transmit short sample video segments to FNI to confirm the video angle and format is suitable to analysis prior to transmitting the complete video file.

Video Format

  • Minimum frame rate: 15 fps
  • Minimum resolution: 800 x 600
  • Duration: 4 to 8 hours
  • Common, non-proprietary video formats are acceptable (AVI, MP4, etc.)

Camera Placement

  • Video data should be collected at a signalized intersection.
  • Data collection during peak periods is recommended.
  • Camera position and orientation should be fixed for the duration of the video. Pan, tilt zoom settings should not be modified; hand-held camera footage should be avoided.
  • Field of view should be clear of obstructions.
  • Minimize glare by aiming camera away from the sun and other light sources such as street lamps.
  • Collect video during daylight hours.
  • Mounting height should be sufficient to prevent vehicles from obstructing one another, approximately 7.5 meters (25 feet) or higher.
  • Distance from camera to opposite side of intersection should be less than 45 meters (150 feet).

Good camera placement

✔ Free of obstructions
✔ Camera adjacent to intersection
✔ Left turn movements within 45 meters (150 feet) of camera

Poor camera placement

✘ Glare obscures opposite approach

Poor camera placement

✘ Traffic light obscures some movements

 
 

Registration


To register to participate, please complete this form.

Registration is open until June 30, 2017 and will function on a first-come, first-serve basis. Registration will closer prior to June 30, 2017 if all spaces fill up sooner.

Please request the ideal number of intersections you would like analyzed. The number of intersections per participant will be finalized at the close of the registration period.

Participants will be scheduled into two groups depending on when they are available to collect and transmit video data to FNI for analysis. Please indicate which group you would like to register under.

Group 1
Deadline to submit video to FNI: August 14, 2017
Receive results on intersection speed study from FNI: October 2017

Group 2
Deadline to submit video to FNI: October 16, 2017
Receive results on intersection speed study from FNI: December 2017

By registering to participate, you agree to:

  • Provide FNI with video data meeting the requirements specified in this document by August 14, 2017 (for Group 1) or by October 16, 2017 (for Group 2).
  • Grant ownership of the video data and study results to FNI. FNI will not delete or return video data after the study has been completed. Granting ownership of the video data to FNI means that:
    • FNI will analyze the video data for the purpose of producing a site-specific speed study report which you will receive. The results of this site-specific study may also be shared publicly by FNI at any time.
    • FNI will utilize the video data for making various software improvements, including but not limited to improved object tracking and classification algorithms, improved safety indicators, and improvements to video processing.
    • FNI will analyze the video data for the purpose of producing case studies, conference publications, and journal publications on vehicle speeds, pedestrian safety, road safety, and video analytics. The results of this research may be shared publicly by FNI at any time.
    • FNI will not publish results containing personally identifiable information contained on video files.

 

For more information on this initiative, please contact us at: 

☎  info@fireseedsnorth.ca

 

Project Timeline and Updates


May 8, 2017

The project is launched to mark the start of the fourth annual UN Global Road Safety Week.

May 26, 2017

Participants from over 20 countries have registered for free video-based Safe Systems speed studies. There are still spots available; to claim one of the remaining spots, please complete the form at the bottom of the page.

Countries with jurisdictions and other organizations participating in free Safe Systems speed studies are indicated in yellow.

Countries with jurisdictions and other organizations participating in free Safe Systems speed studies are indicated in yellow.

June 5, 2017

Registration packages are now ready. Anyone interested in participating in the project who has not already indicated their interest is also asked to fill out the registration package form.