Work

Deep Learning Enhanced Multisensor Data Fusion for Building Assessment Using Multispectral Voxels and Self-Organizing Maps

Year: 2024

Type: article

Abstract: Efforts in the domain of building studies involve the use of a diverse array of geomatic sensors, some providing invaluable information in the form of three-dimensional point clouds and associated registered properties. However, managing the vast amounts of data generated by these sensors presents significant challenges. To ensure the effective use of multisensor data in the context of cultural heritage preservation, it is imperative that multisensor data fusion methods be designed in such a way as to facilitate informed decision-making by curators and stakeholders. We propose a novel approach to multisensor data fusion using multispectral voxels, which enable the application of deep learning algorithms as the self-organizing maps to identify and exploit the relationships between the different sensor data. Our results indicate that this approach provides a comprehensive view of the building structure and its potential pathologies, and holds great promise for revolutionizing the study of historical buildings and their potential applications in the field of cultural heritage preservation.

Source: Heritage

DOI: 10.3390/heritage7020051

Authors: Javier Raimundo, Serafin Lopez-Cuervo Medina, Julian Aguirre de Mata, Tomás Ramón Herrero-Tejedor, Enrique Priego-de-los-Santos

Institutions: Universidad Politécnica de Madrid, Universitat Politècnica de València

Language: English

Topic: 3D Surveying and Cultural Heritage

Subfield: Geology

Field: Earth and Planetary Sciences

Domain: Physical Sciences

SDG: Sustainable cities and communities

Open Access status: gold

Estimated APC paid: $1,819