Improving Rail Safety: How Computer Vision Technology is Reducing Incidents at Level Crossings

Rail level crossings have been a significant safety concern in Australia for many years, with fatal and non-fatal incidents occurring regularly.

Between 2011 and 2020, there were a total of 355 collisions between trains and road vehicles at level crossings in Australia, leading to the deaths of 66 people. 50% of these fatalities occurred in regional areas, while the other 50% occurred in metro areas. The most common type of incident in metro areas was collisions between cars and trains, accounting for 53% of all incidents. In contrast, the most common type of incident in regional areas was collisions between road vehicles and trains, accounting for 67% of all incidents.

The Victorian Department of Transport and Planning (DTP) has been trialling Felicity's Multimodal True Edge Traffic monitoring service, which uses computer vision technology to identify potential hazards and alert drivers, pedestrians, and train operators to take appropriate action. The system is a stand-alone edge monitoring system that can also be retrofitted to existing infrastructure, making it a cost-effective solution to address safety concerns at rail-level crossings. By using real-time data and analysis, the system can detect potential hazards quickly, providing an additional layer of safety for drivers, pedestrians, and train operators.

The data captured by the system includes multimodal dwell time in critical zones, real-time traffic flow, and speed capture. Having a continuous real-time feed of data has uncovered some important statistics that show the dangers around metropolitan rail level crossings.

The data can be pushed through an API to SCATS integration (or other services), which decodes the real-time data from Felicity, defines rules based on the data findings, and activates the SCATS priority engine via a server. This enables real-time SCATS signal cabinet actuation, allowing for rapid response to potential hazards.

The use of computer vision analysis on the edge offers many benefits compared to traditional safety measures, such as warning signs and barriers. By using real-time data and analysis, the system can respond to potential hazards quickly, providing an additional layer of safety for drivers, pedestrians, and train operators. It is particularly effective in regional areas where there is often less awareness and education about the risks associated with rail-level crossings.

Felicity's case study on the DTP rail level crossing in a metropolitan area of Australia highlights the effectiveness of this technology in reducing the number of incidents. By adopting Felicity's Multimodal True Edge Traffic monitoring service, the Victorian DTP is taking proactive steps towards reducing the number of incidents and improving safety at rail level crossings.

To find out more on how you could implement the Felicity Multimodal True Edge Traffic monitoring service contact us.