icon

Asphalt Defect Detection

Transform road pavement assessments by automatically detecting pavement issues and delivering rapid insights to decision-makers.

Beta
Free Trial available!
Asphalt Defect Detection
Supervised Machine Learning
This composite vision model combines two distinct types of machine learning: object detection to detect non-linear defects like patching; and instance segmentation to detect linear defects, such as transverse and longitudinal cracks.
Use Case
Expedite your pavement assessment process with the model’s summary report of defects founds.
Fast Analysis
The model is currently trained to detect:
  • Alligator cracking
  • Longitudinal cracking
  • Patching and utility cut patching
  • Potholes
  • Transverse cracking
Note: This model is intended for beta testing only. It has a limited feature set and should not be used for production applications.

Required Inputs

  • PNG, JPG, or JPEG files

Note: This model is trained on data captured from a vehicle-mounted camera, from the perspective of the front/back of the vehicle, facing down on the road. Submitted images should be oriented toward the direction of the road, not sideways.

Expected Outputs

  • Annotated images
  • A summary report in XLSX format

The Asphalt Defect Detection model streamlines the process of assessing asphalt pavement conditions, eliminating labor intensive, subjective, and expensive evaluation methods. This machine vision model detects and classifies common pavement defects, providing insight on road networks. Optimize your pavement management strategies by quickly identifying asphalt defects in minutes!

Note: Current mAP (Mean Average Precision) score is 50%. Results will continue to be refined as the model is trained on more data.

Cookie Statement

We collect cookies to improve your experience with BRYX.