The purpose of the BIMprove Risk and Image Data Management Service is to store and provide access to the data created by the Safety Related Visual Object Detection described above. The Risk and Image Data Management Service includes three main parts that are:

  1. The database used to store the metadata provided by the Safety Related Visual Object Detection. The selected database solution is Apache Fuseki graph database that supports RDF (Resource Description Framework) format and SPARQL query language. The database structure follows the risk data ontology specifically developed to address the requirements of risk data management. A visualization of the ontology is shown in Figure 3.
  2. REST interface that provides an access to the database. The interface includes two basic access methods that are insert data and query data. The interface is implemented using Python language and Tornado web framework with its asynchronous networking library.
  3. HTML based GUI that enables examining the content of the database (as well as the results of the Safety Related Visual Object Detection). The GUI is supported by the Flask micro web framework written in Python.

Figure 3: Risk ontology

The source code implementing the aforementioned modules is available at

Do you want to know more? Download our deliverable from heresubscribe to our YouTube channel and share your opinion with us through our LinkedIn or our Twitter communities!