The paper presents a predictive framework for monitoring architectural decay by integrating 3D survey data with artificial intelligence techniques. The methodology is structured as a Digital Decision System (DDS) that combines multi-source data acquisition (laser scanning, photogrammetry, mobile mapping, and low-cost sensors) with machine learning models to forecast degradation trends. Following a Knowledge Discovery in Database (KDD) pipeline—comprising data collection, preprocessing, transformation, data mining, and interpretation—the system uses linear regression to model the relationship between temporal data and measurable degradation phenomena. The approach is validated through experimental applications, demonstrating how predictive models can support conservation strategies, optimize monitoring processes, and assist decision-making in architectural restoration.
Evaluation of Annotation Ambiguity in Common Supervised Machine Learning Classification Approaches for Cultural Heritage
The paper investigates the impact of annotation ambiguity on supervised machine learning classification processes applied to cultural heritage data. Focusing on semantic segmentation of 3D point clouds, the study highlights how manual annotations—used as training data—are inherently subjective and influenced by the annotator’s background, expertise, and interpretation of architectural elements. Through an experimental framework involving multiple annotators, the research quantifies variability in labeling using metrics such as Intersection over Union (IoU), revealing significant discrepancies in the definition of semantic classes (e.g., columns, arches, moldings). The work proposes integrating uncertainty-aware approaches, such as semantic maps, to encode annotation ambiguity and improve the robustness, interpretability, and interoperability of AI-based classification systems for heritage documentation.
Sicilian Heritage Identity: Between Stereotype and AI-Based Knowledge
The paper investigates how artificial intelligence can be used to visualize and analyze the collective imaginary of a place, focusing on the representation of Sicily through textual descriptions. By applying text-to-image AI models to literary excerpts, the research aims to make explicit the intangible and often stereotypical mental images associated with the Sicilian landscape. The methodology combines textual analysis of selected novels with iterative image generation, examining recurring visual patterns and the influence of lexical structures on the output. Results show that AI tends to reproduce dominant visual archetypes (e.g., horizon lines, central compositions, recurring elements such as sea, boats, or rural landscapes), while struggling to interpret syntactic relationships and complex semantic nuances. The study highlights both the potential of AI as a tool for exploring cultural perception and its limitations in translating abstract, narrative-based descriptions into coherent visual representations.
AI and XR for the Knowledge, Monitoring and Promotion of Cultural Heritage Places: The Heritour Project
The paper presents the Heritour project, an integrated framework combining artificial intelligence, sensor-based monitoring, and extended reality technologies for the management and promotion of cultural heritage. The approach focuses on the development of a predictive monitoring system based on machine learning algorithms, capable of processing data collected through IoT sensors and integrated surveys to assess structural conditions and potential risks. At the same time, the project leverages XR technologies (VR/AR) to enhance dissemination, enabling immersive and interactive exploration of heritage sites. The workflow integrates data acquisition, processing, and management within a multi-layered digital infrastructure, supporting both decision-making processes and public engagement through semantic 3D models and digital platforms.
AI for Archaeological Heritage Applications
Lo studio propone una metodologia basata sull’integrazione di algoritmi di intelligenza artificiale e algoritmi generativi per supportare la ricostruzione di manufatti archeologici frammentari. Attraverso tecniche di image generation (GAN, Transformer), semantic segmentation e modelli computazionali parametrici, il sistema consente di analizzare frammenti, riconoscerne l’appartenenza tipologica e generare ipotesi ricostruttive tridimensionali. L’approccio viene validato tramite un workflow iterativo che combina dataset visivi, classificazione geometrica e ottimizzazione tramite algoritmi genetici, con verifica finale da parte di esperti.
Artificial Intelligence for Space Weather Prediction
This paper reviews the application of artificial intelligence methods for space weather prediction, with a focus on solar flare forecasting. It contrasts physics-based and data-driven approaches, highlighting how machine learning models trained on large datasets of solar magnetograms and flare records can provide probabilistic predictions. While AI methods represent the current state-of-the-art, challenges remain in terms of data imbalance, feature selection, and model interpretability.
Hybrid AI-Based Annotations of the Urban Walls of Pisa for Stratigraphic Analyses
This paper proposes a hybrid methodology for the semantic annotation and stratigraphic analysis of architectural heritage, combining supervised machine learning techniques with photogrammetric 3D modeling. The research aims to support the interpretation of complex masonry structures by integrating 2D image-based classification with 3D spatial representations.
The workflow involves the acquisition of photogrammetric datasets, followed by supervised classification of images using a Random Forest algorithm trained on manually annotated samples. As illustrated in the workflow diagram (Fig. 2, p. 783), semantic information related to lithotypes and construction layers is first propagated across the entire image set and subsequently transferred to the 3D point cloud through 2D–3D reprojection. This process enables the creation of semantically enriched 3D models, preserving the relationship between geometric data and interpretative annotations.
Results, shown in the classification outputs (Figs. 4–6, pp. 786–788), demonstrate that the method effectively identifies different construction phases and material layers, particularly in cases with clear colorimetric variation. However, limitations arise in areas with homogeneous textures, where classification accuracy decreases. The study highlights the potential of hybrid AI-based annotation systems to enhance stratigraphic analysis and heritage documentation, while emphasizing the need to integrate additional geometric features to improve robustness.
Monitoring Systems Design with Real Time Interactive 3D and Artificial Intelligence
This paper presents a methodological framework for the design and simulation of monitoring systems for cultural heritage, integrating real-time interactive 3D environments and artificial intelligence techniques. The research focuses on preventive conservation, proposing a workflow that combines multiscalar digitization, semantic annotation, and AI-driven analysis to detect and predict the evolution of degradation phenomena.
The methodology is structured into three main phases: (i) detailed digital acquisition of the architectural artifact through integrated survey techniques (TLS, photogrammetry, and image-based methods); (ii) construction of an information-rich 3D model, semantically annotated and enriched with diagnostic data (thermal, material, and colorimetric); and (iii) simulation of monitoring systems within a real-time interactive environment using Unreal Engine 5. As illustrated in the workflow diagram (Fig. 1, p. 722), the system enables the placement of virtual sensors and the generation of synthetic datasets through simulated damage patterns. These data streams are used to train machine learning models for anomaly detection, which are integrated with natural language–based AI systems to support user interaction and decision-making. The results demonstrate the potential of combining digital twins, simulation environments, and AI to design and test monitoring infrastructures prior to on-site implementation, while highlighting the challenges related to data integration, model accuracy, and real-world validation.
Dataspace: Predictive Survey as a Tool for a Data Driven Design for Public Space
This paper proposes a data-driven methodological framework for the analysis, simulation, and design of public space through predictive survey techniques. The research integrates digital survey methods, big data acquisition, and artificial intelligence to construct a three-dimensional model capable of supporting decision-making processes in urban design. The approach is structured into four phases: historical and contextual analysis, morphometric data acquisition through photogrammetry and LiDAR, digital model construction, and predictive simulation.
The study develops a digital twin environment in which heterogeneous data—collected from field surveys, low-cost sensors, and human-generated sources—are integrated and analyzed. As illustrated in the workflow diagram (Fig. 1, p. 705), the system combines physical and digital datasets to simulate future scenarios and evaluate public space quality. The methodology incorporates both quantitative metrics and qualitative assessments based on Gehl’s criteria, enabling the interpretation of spatial use, environmental conditions, and user behavior. The results demonstrate that predictive survey, combined with IoT and machine learning techniques, can support the design of responsive and adaptive urban environments, while highlighting current limitations in data acquisition, integration, and systematization.
AR-Bicycle: Smart AR Component Recognition to Support Bicycle’s Second Life
This paper presents an augmented reality–based system for object and component recognition aimed at supporting product maintenance, repair, and lifecycle extension within a circular economy framework. The research focuses on bicycles as a case study, proposing a mobile application that combines 3D modeling, computer vision, and AR visualization to identify components and link them to contextualized repair information. The methodology integrates geometric analysis, average-shape modeling, and deep learning–based object recognition to enable scalable detection across different bicycle types. The workflow includes the definition of points of interest (PoIs), the creation of a generalized 3D model, and its implementation within Unity and Vuforia environments for AR interaction. As illustrated in the methodological diagram (Fig. 4, p. 613), the system connects recognition, information retrieval, and user interaction through a structured pipeline. The application allows users to visualize repair instructions, access multimedia content, and locate nearby service points, bridging digital knowledge and physical intervention. Results demonstrate the effectiveness of AR in improving component awareness, facilitating repair practices, and promoting sustainable product use, while highlighting limitations related to geometric variability and recognition accuracy across different product typologies.