Enhancing innovation and safety in the rail industry with computer vision

Railway tracks and trains require constant inspection and maintenance.

Infrastructure safety and stability are critically important for this North American Class 1 railroad.

When precious cargo is transported over railways, safe passage hinges on impeccable maintenance. For this railroad, relying on human inspection of its wheels and tracks was a time-consuming, costly, and inaccurate process with the potential for disastrous results.

The railway provider deployed Cogniac to inspect over 22 million wheels and 60,000 miles of track every single month. Today, cameras are installed on 450 trains, photographing the track at 60mph. Those images are sent to the edge and processed by the Cogniac platform, which alerts a human operator in real time whenever a defect or issue is detected.

1.5M wheels a day are inspected for defects, including cracked components or wear and tear.

As onboard sensors and cameras survey the track structure passing beneath, Cogniac detects flaws imperceptible to the human eye.

This data is processed using algorithms and artificial intelligence to determine when and where track maintenance is needed.

Cogniac’s machine learning algorithms analyze the data collected from the images of the tracks and wheels. This data provides valuable insights that can help identify trends, predict potential issues, and make more informed decisions about infrastructure maintenance and management. When a potential issue is detected, the platform immediately flags it for review to determine whether maintenance or repairs are necessary.

The use of machine learning algorithms also enables Cogniac to improve its accuracy over time by continuously learning and adapting to new data. This means that the more data the platform collects and analyzes, the more effective it becomes in identifying potential issues and providing valuable insights.