ReAD is an interdisciplinary research project financed by a Lazio Innova’s €150.000 grant and carried on by the Istituto di Scienze e Tecnologie della Cognizione of Consiglio Nazionale delle Ricerche (ISTC-CNR), the Department of Architecture of Roma Tre University (DArc-RM3) and the Istituto Centrale per il Catalogo e la Documentazione of the Italian Ministry of Culture (ICCD-MiC). The multidisciplinary team includes computer scientists, preservation architects, ontology engineers, economists, anthropologists. ReAD works in the field of architectural heritage and aims to improve its knowledge by using artificial intelligence (AI) – i.e., machine learning and knowledge representation technologies – in the data collection processes. The overall objective is to automate the acquisition of information related to architecture and to link it to data from other sources, increasing the knowledge about immovable heritage thanks to semantic Web. Specifically, ReAD is working on the development of a computational technology that enables the automatic extraction of structured data from unstructured sources relating to the domain of architectural heritage and including natural language text corpora and images. Therefore, this tool will be able to read (as the project acronym says), analyze and process iconographic and textual sources, which will be enhanced, exploited and used in a new way. The images used include pictures and historical drawings and are part of the ICCD’s prestigious collections, i.e., the Catalogo generale dei Beni culturali and the Gabinetto Fotografico Nazionale. The project's aim and methodology are extremely innovative: there are some studies which have applied computer vision to architectural images, but they have mostly worked on pictures and digital CAD drawings. The image recognition applied to historical hand-drawing is a very experimental field of research. In the following steps, this information will be made accessible within the semantic Web, thanks to knowledge graphs and Linked Open Data (LOD). The team is working on modeling open-source ontologies, which will organize the extracted data in a coherent and semantically structured system, able to effectively describe and represent the knowledge of the architectural heritage domain. These ontologies will enrich ArCo, the ontology network of the Italian cultural heritage, which has been developed by ICCD-MiC and ISTC-CNR, in collaboration with University of Bologna and DArc-RM3 (as far as the architectural heritage domain is concerned). The expected outputs (the image recognition tool, the ontologies, the LODs, the architecture dataset) will be released open source and will be available for further developments for the companies and start-ups.

Birrozzi, C., Coco, A., D’Abate, S., D’Amore, F., Frangipane, M.C., Gangemi, A., et al. (2023). ReAD | Representation of Architectural Data. Enhancement of architectural heritage through the application of artificial intelligence. In Diagnosis for the Conservation and Valorization of Cultural Heritage (pp.346-357). Giugliano in Campania : Cervino Edizioni.

ReAD | Representation of Architectural Data. Enhancement of architectural heritage through the application of artificial intelligence

E. Pallottino;P. Porretta;
2023-01-01

Abstract

ReAD is an interdisciplinary research project financed by a Lazio Innova’s €150.000 grant and carried on by the Istituto di Scienze e Tecnologie della Cognizione of Consiglio Nazionale delle Ricerche (ISTC-CNR), the Department of Architecture of Roma Tre University (DArc-RM3) and the Istituto Centrale per il Catalogo e la Documentazione of the Italian Ministry of Culture (ICCD-MiC). The multidisciplinary team includes computer scientists, preservation architects, ontology engineers, economists, anthropologists. ReAD works in the field of architectural heritage and aims to improve its knowledge by using artificial intelligence (AI) – i.e., machine learning and knowledge representation technologies – in the data collection processes. The overall objective is to automate the acquisition of information related to architecture and to link it to data from other sources, increasing the knowledge about immovable heritage thanks to semantic Web. Specifically, ReAD is working on the development of a computational technology that enables the automatic extraction of structured data from unstructured sources relating to the domain of architectural heritage and including natural language text corpora and images. Therefore, this tool will be able to read (as the project acronym says), analyze and process iconographic and textual sources, which will be enhanced, exploited and used in a new way. The images used include pictures and historical drawings and are part of the ICCD’s prestigious collections, i.e., the Catalogo generale dei Beni culturali and the Gabinetto Fotografico Nazionale. The project's aim and methodology are extremely innovative: there are some studies which have applied computer vision to architectural images, but they have mostly worked on pictures and digital CAD drawings. The image recognition applied to historical hand-drawing is a very experimental field of research. In the following steps, this information will be made accessible within the semantic Web, thanks to knowledge graphs and Linked Open Data (LOD). The team is working on modeling open-source ontologies, which will organize the extracted data in a coherent and semantically structured system, able to effectively describe and represent the knowledge of the architectural heritage domain. These ontologies will enrich ArCo, the ontology network of the Italian cultural heritage, which has been developed by ICCD-MiC and ISTC-CNR, in collaboration with University of Bologna and DArc-RM3 (as far as the architectural heritage domain is concerned). The expected outputs (the image recognition tool, the ontologies, the LODs, the architecture dataset) will be released open source and will be available for further developments for the companies and start-ups.
2023
978-88-95609-82-9
Birrozzi, C., Coco, A., D’Abate, S., D’Amore, F., Frangipane, M.C., Gangemi, A., et al. (2023). ReAD | Representation of Architectural Data. Enhancement of architectural heritage through the application of artificial intelligence. In Diagnosis for the Conservation and Valorization of Cultural Heritage (pp.346-357). Giugliano in Campania : Cervino Edizioni.
File in questo prodotto:
Non ci sono file associati a questo prodotto.

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11590/463389
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact