SEEBRIDGE - Automated Compilation of Semantically Rich Bim Models of Bridges

SeeBridge (Semantic Enrichment Engine for Bridges) targets the development of a comprehensive solution for rapid and intelligent survey and assessment of bridges. Various advanced remote sensing technologies, including terrestrial laser scanner, high-resolution cameras mounted on mobile vehicle and drone, etc., will be used to rapidly and accurately capture the state of more than 5 bridges across the world and produce high density coloured point cloud data.

A bridge object detection software will be developed to recognize distinct 3D solid geometry from the point cloud data. An expert system will be developed for classification of bridge components from the 3D solid model and for deduction of supplementary information concerning material types, internal bridge component, etc., using encoded bridge engineers’ knowledge. A damage measurement tool will also be devised to associate the identified defection to the model at the bridge component level.

The output will be a Bridge Information Model that is semantically meaningful to sufficiently provide most of the information needed for decision-making concerning the repair, retrofit or rebuild of a bridge.

Expected deliverables:

1. Information Delivery Manual (IDM) for typical information requirements for generic Bridge Management System (BMS) and for repair/rehabilitate/retrofit of highway bridges
2. Integrated processing tool for automated compilation of 3D solid bridge model objects from point clouds (w/registered imagery), including working prototype and demonstration
3. A semantic enrichment engine
4. A series of rule-sets for inferring bridge components based on 3D geometry and topological relationships
5. An expert system for semantic enrichment of bridge BIM models
6. Graphical user interface and annotated dataset of images for damage detection and classification
7. An integrated system for automated damage detection and classification