Traffic Datasets Obtained By Video Analytical Surveys On New “post-COVID” Schemes
Published by Daniel Stofan on
Many local authorities are considering more long-term moves to restrict car use in the wake of the COVID-19 crisis. Figures from local governments in the UK are showing that local authorities are looking to take advantage of the crisis to push for environmental and sustainable travel plans. They are looking at reclaiming the existing road space away from road vehicles and giving it to those sustainable travel modes like bicycle and foot traffic.
Social distancing measures are compatible as long as people stay at least two meters apart. You can measure that, but what are the findings besides that not all people adhere to it? This highlights the lack of safe space in some areas to allow people to make essential trips and exercise in the safest possible way. Therefore, the new measures the cities are undertaking are aimed to help citizens to stay healthy under the changing circumstances and to keep the traffic efficient in the new normal.
New schemes are being created, temporary or permanent bike lines are added, well-trafficked streets are becoming one-ways, roads are being closed temporarily, etc. How is this helping the purpose and how can we evaluate if the traffic infrastructure measures are efficient? This is the question that can be answered by smart observation of the traffic before and after the measures are taken.
There was much written about the modern software traffic analytical systems and traffic data collection. Today the benefits of this technology are becoming more and more obvious and its flexibility and speed of deployment comes handy in the rapidly changing environment. This article describes some of the useful traffic parameters and patterns that can be analysed from the traffic video footage when doing the surveys of the new traffic schemes. If you’re new to automated traffic analyses using the artificial intelligence, check out some of our previous articles or visit www.goodvisionlive.com to learn more.
A most common analysis is capturing the traffic volumes on streets and sidewalks. Traffic counting over the digitally collected data from cameras can be performed interactively over custom-defined traffic movements. GoodVision provides a classification of traffic attendants into 8 providing the multi-modal counts on each turn, regardless of the type of the junction. Every traffic obtained by GoodVision can be visualised and exported into Excel reports.
Traces visualization and heatmaps
GoodVision extracts and visualises full object trajectories with exact positions in time with millisecond granularity. This is amazing not only for visual analytics but also as the hi-quality baseline for calculation of all further traffic metrics. Se the vehicle trajectories sample on the image below:
Goodvision trajectory map displays all individual traces of all traffic attendants
GoodVision visualises the volumes of traffic on the scene via heat-maps, as well as the areas with reduced speed, braking or acceleration.
The versatility and the trajectory nature of the digitally collected data in GoodVision allow to analyse all the physical aspects of traffic, or visually inspect the situation. You can visually inspect violations, lane changes, red-light running, jaywalkers, lane-jumpers and many more.
Advanced traffic parameters
Digital representation of data in GoodVision makes it possible to obtain and report the advanced behavioural parameters of traffic attendants such as speed, travel times, occupancy time and delays.
Travel time and speed
GoodVision provides you with precise travel time information for any traffic attendant passing between detection gates on the junction. Travel time can be reported as the average value for the whole traffic flow, or only for a specific class of the vehicles, or each individual object in the stream. GoodVision provides the estimation of vehicle speed on any turning movement, including the roundabout passages.
Time-gaps (headway time)
valuable information to understand traffic flow behaviour, are needed for calculation other parameters like saturation flows, etc. Time-gaps can be reported as the average value for the whole traffic flow, or each object in the stream, for example, to identify which vehicles are leaving unutilised space on the highway lanes.
GoodVision allows you to analyse headway times between individual vehicles in each lane separately
Occupancy and delay
very important when analysing waiting times on lights, waiting time in traffic queues or measuring the time spent on yielding on turns. GoodVision provides average occupancy time, occupancy time of each object, or the occupancy time heatmap
GoodVision detects stopped vehicles and allows you to use this event as a filter. If you were ever wondering how many vehicles stop/not on a STOP sign, clearway sign or how many vehicles are waiting/not on red lights and then turn or go straight, this is the solution.
License plates and origin-destination analysis
Similarly to O-D monitoring on the micro-level, GoodVision can provide the origin-destination studies in a way more significant scale. Using our ANPR capability, you can perform the O_D studies, obtaining license plate numbers and getting the travel counts and journey times from the bigger network.
Bicycle and pedestrian monitoring
With bicycle measures becoming more and more popular, it is essential today to update the traffic schemes with this modality. All the traffic analyses described above and below are available also for the study of bicycle and foot traffic, making it very suitable for the reviews of new city schemes with measures taken in favour of these modalities.
GoodVision captures bicycles and pedestrians with amazing accuracy
Data for traffic simulation tools
There are many parameters that we calculate for their traffic models in simulation systems like PTV Vissim or Linsig, like saturation flows, detecting the ideal free-flow conditions, discovering the areas with reduced speed, entering the traffic O-D volumes to their models directly, and many more.
an essential measurement of on-street performance. Traditionally, Sat Flows were calculated manually for each traffic lane. Now GoodVision provides fully automatic calculation of saturation Flows into the Excel report for your traffic model verification.
Exporting the model to PTV Vissim
A revolutionary functionality in GoodVision Video Insights! Export of the default traffic control model for your simulation system directly to PTV Group Vissim from the video-extracted data on our platform. Just imagine: by importing the pre-calibrated model, your default network will be immediately created and placed on your map in Vissim, saving you tons of hours. The model network exported from GoodVision Video Insights is already pre-calibrated by all necessary traffic parameters for all traffic movements and lanes:
- Traffic volumes per each time-interval
- Average travel times and speed per movement
- And the average time gap between vehicles
Model obtained from Video Insights:
- Is placed directly to the Vissim background map
- Contains links and their real distances
- Contains links and route decisions in the same way as modeller would draw manually
- Is by default calibrated by multimodal vehicle counts and all necessary traffic parameters
Try GoodVision Video Insights in a Free Trial
If you feel frustrated or intimidated by all that tech-talk and variety of vendor lock-in single-purpose solutions out there, then GoodVision is the way to go. If you need to process just a few or thousands of hours of traffic videos, Video Insights can handle it. It is built to scale even for your massive projects and to deliver traffic data always in 1 hour. Video Insights is a broad ecosystem of capabilities and a one-stop-shop for all the traffic analytical tasks the transportation industry requires. Our goal is to remove the stress from your traffic jobs with automation and conserve your time for professional tasks!
Start your trial at my.goodvisionlive.com or contact our consultants at email@example.com to learn more and get a walkthrough of the platform.