AI Road Condition Assessment
Run deep learning-based Gravel Road Assessment Models on your dashcam video. They detect and locate issues such as rutting and dirty or muddy road, and assign a road condition class, so you can rate your gravel road network without a windshield survey.
Rating hundreds of kilometres of gravel road by eye is slow and subjective. MRF AI Road Condition Assessment uses deep learning-based Gravel Road Assessment Models that run on dashcam video. They detect and locate issues on the road surface and classify the road’s condition, so every section is assessed the same way.
The results link with the dashcam footage they came from, so staff can review any detection in context. They also link with MRF Asset Management, where condition classes feed gravel road ratings and maintenance planning. For panoramic road video, see 360 Road Video.
Above: The models outline and label issues such as rutting and dirty road in dashcam footage, with a confidence score for each.
What you can do with AI Road Condition Assessment
Built for gravel roads
Deep learning-based Gravel Road Assessment Models, trained specifically on gravel road conditions.
Runs on dashcam video
Use the dashcam video your fleet already records. The models process the footage frame by frame.
Detect issues
Each issue the model finds, such as rutting or a dirty road, is outlined and labelled with a confidence score.
Locate every issue
Detections are tied to GPS position from the video, so each issue lands at the right place on your road network.
Road condition classes
Each road section receives a condition class, giving you a consistent, network-wide view of gravel road condition.
Straight into asset management
Results flow into MRF Asset Management to support gravel road ratings, maintenance and rehabilitation planning.
Works well with
See it working with your own data.
Book a demo and we’ll walk you through the products that fit your departments, configured for organizations like yours.