Human-auditable pavement crack assessment: multi-task CTAG-Seg inference, Grad-CAM explainability, transparent PCI & local reports.
-
Updated
Jul 14, 2026 - Python
Human-auditable pavement crack assessment: multi-task CTAG-Seg inference, Grad-CAM explainability, transparent PCI & local reports.
YOLO pipeline detecting 13+ road-damage classes from vehicle footage, GPS-tagging each defect via GNSS fusion
Official A4R Infra website – infrastructure, highway construction, asset management, operations & maintenance, and project information.
This repository contains the design, hardware schematics, software code, and real-world images of our pothole filling robot. The robot is equipped with a Raspberry Pi 4B microprocessor for control, servo motors for efficient gravel mix dispensing, and a sturdy chassis for navigating various terrains.
Measures pothole volume from one photo: YOLOv8-Seg finds the hole, Depth Anything V2 reads how far the surface drops, and a card or coin in frame supplies the scale. Streamlit dashboard with severity, repair cost, GPS mapping, and PDF reports.
FYI Help Here — ASP.NET Core MVC road maintenance reporter app. Report potholes, debris and road hazards to city departments.
To associate your repository with the road-maintenance topic, visit your repo's landing page and select "manage topics."