Within a local authority’s road maintenance department, an agent regularly received reports of road damage from both the public and field agents. The multitude of reports and the diversity of information sources made rapid identification of at-risk areas and efficient transmission to the relevant teams particularly difficult, whilst also requiring thorough follow-up.
DINA facilitates this work by analyzing images from the field:
- Automatic detection of visible anomalies (potholes, cracks, deformation, etc.) from photos or videos
- Precise geolocation of anomalies
- Generation of a contextualized alert with the image, location, and nature of the defect
- Direct transmission to the relevant department with possible prioritization based on level of urgency
- Simplified intervention tracking through image and action indexing
- Seamless integration with existing intervention management tools
Benefits:
- Time savings in incident detection and transmission
- Reduced intervention times and improved risk prevention
- Complete traceability from detection to resolution
- Agents focused on organizing repairs, with reliable information from the outset
DINA incorporates computer vision models trained to automatically detect visible road damage (potholes, cracks, subsidence) from images or videos, regardless of the reception channel (mobile application, email, civic platform). Each anomaly is geolocated via metadata or embedded GPS, enriched with an automated diagnosis, and then transmitted to the competent department via API or business connector. The alert is indexed in the intervention system to ensure complete follow-up, from detection to resolution, while guaranteeing seamless integration with existing GIS, CMMS, or ticketing tools.