TrafficSurvey – Video post-processing platform

Let's meet!
Intertraffic 2024
16 - 19 April 2024
Amsterdam / Netherlands
Traffex
22 - 23 May 2024
Coventry / England
Let's meet!

Application Case - T - junction in Zlin

There are many crossroads where fixed traffic signaling does not match the changing composition of traffic. Suboptimal and poorly executed traffic management leads to traffic jams and driver discomfort. Furthermore it drastically increases emissions and noise and also increases the risk of a traffic accident. This was no different at the T junction in the Zlin city where due to the construction of a new shopping center, the traffic demand rapidly increased and became a lot less predictable. Due to this, the installation of sensors for adaptive traffic control was supported to go ahead under the following conditions:

  • installation without road surface disruption
  • installation without having to stop traffic
  • utilization of existing traffic controller
  • low purchase and maintenance costs
TrafficXRoads dashboard printscreen

Project details

  • Use case: Traffic flow improvement
  • Product: TrafficXRoads
  • Location: Czech Republic (Zlin)
CZE Zlin Sokolska CROSS 3 cameras v.2

Camera installation – detection up to 80-120 meters, multimodal traffic data

TrafficXRoads - adaptive traffic control setup

 TrafficXroads together with cameras is very easy to deploy to existing crossroads and allows to upgrade them with adaptive traffic control capabilities. The traffic engineer responsible for the project required the detection of vehicles at the minimum distance of 50 meters in every direction. There was no cabling between the traffic controller and the individual traffic light poles that had only permanent power supply. Because of this all the data communication was done wirelessly. On each of the traffic light poles one FULL HD IP camera with WDR technology was installed; in total 3 cameras. The TrafficXRoads unit was installed inside the case of the traffic controller and was connected to the ctonroller via an UTP cable. The hardware installation took 3 hours including the verification of the communication and the correct functioning and was done without the need to stop the traffic.

HW was installed the configuration of the sensoric system using the FLOW Insights tool was done. This could have also been done on the spot but thanks to the internet connectivity of the traffic controller a remote configuration was done instead. After connecting to the TrafficXRoads units using the FLOW Insights configuration application, all three camera streams were added and nine detection gates and zones simulating the induction loops were defined based on the prepared designs.


Adaptive traffic control in Zlin

Camera installation – one camera per pole – each camera is able to detect incoming vehicles

Data analytics configuration

Next, the unit was configured for collection of traffic data and monitoring of the effects of adaptive traffic control. For this purpose an interactive dashboard was created in FLOW that was analyzing multiple values and performance measures such as:

  • Number of vehicles that have stopped / total number of vehicles
  • Average stationary duration in the crossroad
  • Traffic intensity – categorization into 7 classes
  • Average vehicle speed

The configuration of TrafficXRoads unit using FLOW INSIGHTS took only 15 minutes for the whole intersection and while the configuration was being done it was possible to see the real-time output of the analysis and collect the traffic data.

Prinstscreen TrafficXRoads setup night

FLOW interface – virtual induction loops setup / configured in a few minutes.

Results

The system is now running over 4 months without a single disruption. The dynamic control of traffic has almost halved the waiting time in comparison to the previously used fixed traffic signaling. Thanks to the system detecting cars up to 80 meters, the number of vehicles having to stop at the intersection has reduced 1.7x times. This has a positive effect on the reduction of the local noise and emission levels. This solution was installed and configured in just a single morning and there was no need to stop the traffic. This solution is considerably cheaper in comparison to other technologies.  

Want to read more? Go back to the BLOG.

Challenge

  • Full coverage – all directions 
  • Detect incoming vehicles at distance of least 80 meters 
  • Reliable traffic monitoring > 98 % in detection, > 90 % in classification
  • Budget constraints  – no construction works / installed as an add-on 
  • No data/cable infrastructure ready

Key features

  • Real-time detection and classification of all traffic participants
  • One camera directional coverage of up to 80 meters
  • Full data integration with traffic controller CROSS via UDP
  • Queue and blocking vehicle detection 
  • Instant traffic statistics for the smart city platform

Results

  • Adaptive traffic signal control – 2x shorter average waiting time compared to fixed signals
  • Reduction of traffic jams at rush hours
  • 24/7 real-time traffic monitoring with camera surveillance 
  • Non-intrusive and fast installation of the system without the need to interrupt traffic 
  • Cost efficiency – replacement of 9 inductive loops by 3 cameras 
  • On-site processing – fully autonomous adaptive traffic control system
  • Fully functional in 3.5 hours

Like the story? Share it!