From a series of digitisation projects at i3mainz arose the need for the standard-compliant provision of metadata for digitised material.
For the assessment of the quality of a digital copy of cultural heritage objects, the complete, machine- and human-readable description of the acquisition and data processing is central. Therefore, i3mainz and RGZM staff are working on a cross-project metadata schema and ontology model for 3D acquisition and data processing.
Based on the expertise and work in the projects ARS3D, Haft Tappeh and Tomb Monuments staff members of i3mainz and the RGZM developed solutions for the technical metadata that arise when using a strip light projector from GOM (a ZEISS company) and the associated scanning and processing software.
The GOM software offers the possibility of accessing technical metadata via a Python API, i.e. a programming interface. In this way, information such as the size of the calibrated measurement volume, settings for processing the individual measurements or residuals of the transformations (deviation values) can be accessed. Python scripts were developed for the GOM software Professional 2016 as well as for a predecessor software ATOS, which read out and structure all information. In a parallel developed ontology, the description of the individual metadata and the mapping to existing standards took place. The ontology standardises the description of the relationships between the quality parameters. The export of the technical metadata is done in the file formats JSON and TTL.
To check the metadata generated in TTL format and the metadata schema set up, a prototype SPARQL database was set up in which the TTLs from different documentation projects were imported. Queries were used to validate the correctness of the imported information about the sensors used, number of scans, resolution, etc. in the database. The aim was not only to standardise the metadata but also to provide a database for future quality analyses on scan data.
In the project year 2021, the ontology and the exported metadata were adapted in the data formats JSON and TTL, application examples of the ontology were prepared and published on Zenodo, the Python software for generating the metadata was released on Github and the overall process of metadata generation was described in a publication documented in detail in the Heritage Science Journal.