Through the use of AI video analytics the City of Bellevue is partnering with Brisk Synergies to achieve their Vision Zero goals.
The city's existing traffic cameras have been used to analyze near misses that occur between all road users. These insights provide Bellevue with more context and data, allowing the city to take more proactive approaches in making their roads safer to achieve their Vision Zero goals.
This strategic partnership with Bellevue is a continuation of work that was initiated by Microsoft.
Network Screening Tool for Identification of Hotspots
Based on the collision prediction models and methods proposed in the HSM, a network screening tool will be integrated into the platform. Each intersection will receive a Brisk Safety Score based on the different Brisk Synergies risk indicators. This tool will help rank and identify the most dangerous intersections.
Employing Machine Learning Models to Identify Risk-Contributing Factors
Performing pedestrian and cyclist cross-sectional safety studies between various intersections in the city, and before/after studies for any high-injury network locations that experienced changes in intersection geometry or signalization during the observed study period.
High Injury Network Studies
Perform deep statistical regression analysis to study high-risk intersections stemming from various traffic conditions. Root causes such as speed, volumes, type of traffic movement, type of interactions, time of the day, road user type, and road geometry are identified.
Complete Traffic-Flow Analysis
At the intersection level, Origin-Destination (O-D) matrices of traffic movement counts (TMC), along with road user classifications, will be provided for all movements captured by the camera and system.
Video-based Network-wide Conflict and Speed Analysis to
Support Vision Zero in Bellevue (WA) United States
Transoft Solutions, the City of Bellevue, Washington and the Together for Safer Roads coalition have partnered on a first-of-its kind, citywide analysis of traffic camera video with the goal of improving safety for all road users. Data collected from over 5,000 hours of video footage were used to produce three reports: 1) Video-based Network-wide Conflict Analysis 2) Video-based Network-wide Speed and Speeding Analysis and 3) Video-based Conflict, Speeding, and Crash Correlation together with an Executive Summary.