AI Portal
For both videos that have been captured live and pre-recorded, this solution provides highly customisable visualisation for traffic and data insights.
&w=64&q=100){kind=small}
)
Tracsis provides Vision-based Survey technology for traffic counting. With swift deployment, accurate analysis, and bespoke reporting we are market leaders.
We use the most advanced digital equipment with machine learning algorithms that allow us to analyse complex mixed traffic situations including simultaneous road, cycle lane and footpath coverage. This AI solution enables video footage of any site to be post-processed in order to accurately capture the volume and speed of all vehicle classes, pedestrians and cyclists.
Advanced Video Analytics
By utilising the same AI technology seen in a permanently fixed AI road sensor and applying it to post-processing of video footage from temporary and ad hoc installations, Tracsis has a new offer that could tackle a gap in the data market as well as address potential limitations to standard data capture.
How it works
The AI software detects, classifies and tracks road users in the video It draws two different speed lines across the road environment to define the area of speed measurement The software uses the real-world distance between the two lines and the width of the road to make its calculations By mapping detected locations into the real-world coordinate frame, the speed of each individual object is obtained.
Safer and more versatile
This method is much safer than traditional alternatives and can also be used where tube installation isn't possible, e.g. high-speed roads. That's because there's no need to enter the live carriageway. Instead, cameras can be safely installed from a road side position.
As accurate as tube counters
After numerous tests on residential, trunk and rural roads, we found the AI Portal to be within .095% accuracy in comparison with a pneumatic tube counter.
Additional benefits
Post-process existing footage from previous projects and provide speed data The speed of active travellers can be captured, e.g. pedestrians and cyclists Even more insights via the AI Portal, e.g. path traces and social distancing data.
An innovative way of visualising the volume and direction of vehicle and pedestrian traffic that's been captured during any video survey work.
Enhanced visualisation and insights
As an enhancement to traditional video footage analysis Tracsis offer a unique path tracing service, offering users a highly customisable way of visualising the volume of vehicle and pedestrian traffic captured during any video survey work.
How it works
Using a proprietary machine-learning system, objects passing through each camera's field of vision are classified into one of 9 categories. Their routes are displayed via trace lines overlaid onto a background image from the relevant camera. The system is highly flexible and traces for each camera are viewable from anywhere between 1 minute to the full duration of the video. The results of individual classifications can be toggled on or off. Crucially, this functionality can be applied to any video survey undertaken by Tracsis.
AI object proximity
This method is much safer than traditional alternatives and can also be used where tube installation isn't possible, e.g. high-speed roads. That's because there's no need to enter the live carriageway. Instead, cameras can be safely installed from a road side position.
Providing insights whilst protecting identities
We extract instant object proximity data, whilst still protecting personal identities. AI analytics processes video footage from our own cameras or other CCTV video feeds.
Additional insights
Interaction Ratio
% of interactions less than 2m (or defined parameter)
Average distance per hour/day
The average duration of interactions <2m Trend analysis