The digitization of graph-based engineering diagrams like P&IDs or circuit drawings from optical sources as well
as their subsequent processing involves both image understanding
and semantic technologies. More precisely, after a raw graph has
been obtained by an object detection and line extraction pipeline,
semantic gaps (like resolving material flow directions) need to
be overcome to retain a comprehensive, semantically correct
graph. Likewise, the graph representation often needs to be
altered to achieve interoperability with established CAE systems
and to accommodate customer-specific requirements. Semantic
technologies provide powerful tools to manipulate such data but
usually require rather complicated implementation. Graphically
presentable graph based rules provide a code-free mean to ease
the interaction with domain experts. In order to be applicable
in real-world applications, both geometric and computational
aspects need to be considered. This paper explores these aspects
and demonstrates use cases of such rule graphs.
| Identifer | oai:union.ndltd.org:DRESDEN/oai:qucosa:de:qucosa:79818 |
| Date | 07 July 2022 |
| Creators | Bayer, Johannes, Li, Yakun, Marquardt, Sebastian, Dengel, Andreas |
| Source Sets | Hochschulschriftenserver (HSSS) der SLUB Dresden |
| Language | English |
| Detected Language | English |
| Type | info:eu-repo/semantics/publishedVersion, doc-type:conferenceObject, info:eu-repo/semantics/conferenceObject, doc-type:Text |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | urn:nbn:de:bsz:15-qucosa2-798165, qucosa:79816 |
Page generated in 0.0016 seconds