The project aims to develop a tool for mapping the degradation state of historical buildings' facades, using AI to analyze data from photographs and thermal images. This initiative represents a significant advancement in the field of heritage preservation, where the need for innovative and effective tools is critical. By employing computational vision to extract information from photographs and thermal images and integrating this data into machine learning models for a comprehensive analysis of the degradation state, the project sets a new standard in the preservation and management of historical buildings.
Work packages:
WP1 - Project management - IPV - 36 months
WP2 - Information management T2 and modelling - FEUP - 12 months
* establish a protocol for the facade's inspection
* create a systematic classification system to organize and document the data
* create the digital twin of the 2 case studies
WP3 - Degradation models - UA - 18 months
* comprehensive lab characterization of the materials and their degradation behavior under controlled conditions
* conduct the in-situ assessment of the case studies
* validate, refine, and improve existing degradation models
WP4 - Tool development - FEUP - 14 months
* combine computer vision and machine learning models to identify, delimit and classify the facades' anomalies in photographs and thermal images
* integrate the new tool in the digital twin
WP5 - Tool implementation - IPV - 15 months
* create protocols for the use of the tool and link it with the predictive maintenance of the building
* test and validate the performance of the tool in real-world environments
WP6 - Communication, dissemination and exploitation - IPV - 36 months |