The increasing adoption of laser scanning technologies, computational methods, and Artificial Intelligence in 3D surveying and architectural digitisation has highlighted the need for automated approaches capable of managing and interpreting large point cloud datasets in support of Heritage Building Information Modelling (HBIM). This research presents the preliminary stage of a Scan-to-BIM workflow focused on point cloud optimisation and segmentation of the ISES social housing complex in Via Smaldone, Foggia (Italy). The proposed methodology includes preprocessing steps aimed at reducing dataset complexity while preserving the geometric and morphological characteristics of the building. Several subsampling and noise-removal strategies were assessed to improve data quality and enhance subsequent computational tasks. The comparison showed that local geometric filtering techniques provide better preservation of façade details and geometric discontinuities than statistical approaches. After the preprocessing stage, an automated segmentation procedure was developed in Python using the open-source library Open3D. The workflow employs an iterative Random Sample Consensus (RANSAC) algorithm to identify and extract dominant planar surfaces directly from the surveyed dataset. The implemented method enables the recognition of the main vertical façade components while excluding secondary elements and irregular geometries. The results demonstrate the effectiveness of the proposed methodology in simplifying complex datasets and supporting the semantic interpretation of architectural features. The segmented point clouds provide a reliable basis for future HBIM modelling and morphological investigations, highlighting the potential of integrating automated extraction techniques and open-source computational tools for built heritage applications.

Point Cloud Optimisation for Scan-to-BIM / Buldo, M., Rossi, N., Verdoscia, C.. - In: INZYNIERIA MINERALNA. - ISSN 1640-4920. - 2:1(2026). (11th World Multidisciplinary Congress on Civil Engineering, Architecture, and Urban Planning (WMCCAU 2026) 31 agosto - 4 settembre 2026) [10.29227/im-2026-02-01-121].

Point Cloud Optimisation for Scan-to-BIM

Buldo, Michele
;
Verdoscia, Cesare
2026

Abstract

The increasing adoption of laser scanning technologies, computational methods, and Artificial Intelligence in 3D surveying and architectural digitisation has highlighted the need for automated approaches capable of managing and interpreting large point cloud datasets in support of Heritage Building Information Modelling (HBIM). This research presents the preliminary stage of a Scan-to-BIM workflow focused on point cloud optimisation and segmentation of the ISES social housing complex in Via Smaldone, Foggia (Italy). The proposed methodology includes preprocessing steps aimed at reducing dataset complexity while preserving the geometric and morphological characteristics of the building. Several subsampling and noise-removal strategies were assessed to improve data quality and enhance subsequent computational tasks. The comparison showed that local geometric filtering techniques provide better preservation of façade details and geometric discontinuities than statistical approaches. After the preprocessing stage, an automated segmentation procedure was developed in Python using the open-source library Open3D. The workflow employs an iterative Random Sample Consensus (RANSAC) algorithm to identify and extract dominant planar surfaces directly from the surveyed dataset. The implemented method enables the recognition of the main vertical façade components while excluding secondary elements and irregular geometries. The results demonstrate the effectiveness of the proposed methodology in simplifying complex datasets and supporting the semantic interpretation of architectural features. The segmented point clouds provide a reliable basis for future HBIM modelling and morphological investigations, highlighting the potential of integrating automated extraction techniques and open-source computational tools for built heritage applications.
2026
11th World Multidisciplinary Congress on Civil Engineering, Architecture, and Urban Planning (WMCCAU 2026)
Point Cloud Optimisation for Scan-to-BIM / Buldo, M., Rossi, N., Verdoscia, C.. - In: INZYNIERIA MINERALNA. - ISSN 1640-4920. - 2:1(2026). (11th World Multidisciplinary Congress on Civil Engineering, Architecture, and Urban Planning (WMCCAU 2026) 31 agosto - 4 settembre 2026) [10.29227/im-2026-02-01-121].
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/11589/310028
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
  • OpenAlex ND
social impact