Bits, Beams & Beauty: The INFN-CHNet Conference on Digital Tools for Cultural Heritage

5 Mar 2026, 13:30 → 6 Mar 2026, 16:00 Europe/Rome

Aula Magna di Ostia (Roma Tre)

Alessandro Re (Istituto Nazionale di Fisica Nucleare), Lisa Castelli (Istituto Nazionale di Fisica Nucleare), Luca Tortora (INFN Roma Tre), Massimiliano Clemenza (Istituto Nazionale di Fisica Nucleare), Valerio Graziani (Istituto Nazionale di Fisica Nucleare)

Bits, Beams & Beauty is the first INFN–CHNet conference dedicated to digital tools and methodologies for cultural heritage, grounded in the scientific expertise developed within CHNet, the INFN network devoted to research on cultural heritage. The conference brings together researchers working across physics, chemistry, materials science, and data science.

The conference will showcase state-of-the-art research covering a wide range of topics, including advanced diagnostics and instrumentation, materials characterisation, artificial intelligence and machine learning, data analysis, and data management for cultural heritage applications. Emphasis will be placed on end-to-end workflows, connecting experimental techniques and data acquisition (“beams”) with digital processing, interpretation, and intelligent tools (“bits”), in support of the understanding, conservation, and valorisation of cultural heritage (“beauty”).

By leveraging the interdisciplinary expertise developed inside the CHNet network, the conference aims to foster collaboration, knowledge exchange, and strategic discussion on current challenges and future directions in digital heritage science. The programme will include contributed talks, poster sessions, and dedicated discussion moments, offering a comprehensive overview of ongoing activities and emerging perspectives within INFN–CHNet and the wider heritage science community.

Organized by 

                 

 

Thursday 5 March

Opening ceremony: Opening

Conveners: Francesco Taccetti (Istituto Nazionale di Fisica Nucleare), Dr Luca Tortora (INFN Roma Tre)

Bits

Digital tools for Cultural Heritage

Convener: Lisa Castelli (Istituto Nazionale di Fisica Nucleare)

1 Digital Services for Cultural Heritage: From Competence Center to Digital Twins

The digital transformation of cultural heritage (CH) has accelerated in recent years, driven by the need for advanced tools to preserve, manage, and valorize historical assets. Within this context, INFN Cultural heritage network participated, trough its members, to a set of initiatives aimed at promoting and valorizing digital competences.
Among such initiatives the 4CH project, funded under Horizon 2020, laid the foundation for a European Competence Centre for the Conservation of Cultural Heritage, aiming to harmonize technological solutions and provide a sustainable infrastructure for CH preservation.
A cornerstone of 4CH has been the development of a cloud-native platform, hosted at INFN-CNAF, designed to deliver scalable and secure services to the CH community. This platform leverages container orchestration (Kubernetes) and federated authentication (INDIGO-IAM) to enable interoperability and resilience.
Beyond infrastructure, 4CH demonstrated its societal impact through initiatives such as SUM – Save the Ukrainian Monuments. Launched immediately after the Russian invasion of Ukraine, SUM aimed to safeguard the digital documentation of Ukrainian cultural heritage at risk of destruction.
The operation mobilized vast resources to store high-resolution images, ensuring that descriptions, photographs, and 3D models of monuments and sites remain available for future restoration and reconstruction, exemplifying how digital technologies can mitigate the consequences of armed conflict on cultural assets.
Building on these achievements, the ARTEMIS project represents the next frontier in CH digital services. ARTEMIS introduces digital twins as a transformative approach to heritage conservation, enabling predictive modeling and simulation of degradation processes and disaster impacts. By integrating theoretical knowledge, existing software, and real-time data into a unified framework, ARTEMIS leverages cloud-based technologies to turn standalone applications into interoperable services, paving the way for advanced modeling capabilities that can replace or complement physical interventions.
The talk will summarize the above-mentioned activities, trying to define possible new challenges and collaborations.

Speaker: Alessandro Costantini (Istituto Nazionale di Fisica Nucleare)

BBB2026_Costantini.pdf

2 Unveiling hidden paintings in pictorial artworks during pre-processing of MA-XRF raw data using fuzzy Gustafson-Kessel clustering with GPU acceleration

Spectral imaging techniques, such as Macro Area X-RayFluorescence (MA-XRF), are extremely powerful methods to investigate the composition of pictorial artworks. The produced data contains a vast amount of information about the composition of the pictorial artwork under scrutiny; nevertheless, post-processing analyses to extract such meaningful information are usually complex, lengthy, and require days (if not weeks) of work by experienced heritage scientists.
A possible approach to tackle these issues is to apply Machine Learning methods in order to extract meaningful information from spectral imaging raw data. In this contribution, we explore the use of unsupervised methods for a fast analysis of MA-XRF raw data. The idea is to see the single-image MA-XRF datacube, i.e. a tensor of shape (𝐻,𝑊, 𝐷),where 𝐻 is the number of pixels in the image height, 𝑊 is the number of pixels in the image width, and 𝐷 is the number of energy bins, as a dataset of spectra of shape (𝑁 = 𝐻 · 𝑊, 𝐷).On this dataset, we perform Principal Component Analysis (PCA), to extract a lower-dimensional representation of the spectra, and then we apply the fuzzy Gustafson-Kessel clustering algorithm to assign to each sample point (i.e .each pixel) a weight, which loosely represent the probability that each pixel belongs to a certain cluster. To speed up the algorithm, we used the JAX framework to move the computation from CPU to GPU, reducing by at least an order of magnitude the execution time (from minutes to seconds).
By rearranging the weight matrix into a set of single-channel images, we get a grayscale image, we can extract visual information in the form of (soft) semantic segmentation. The fuzzyness of the clustering, i.e. the fact that we have grayscale images and not binary maps out of the clustering process, gives additional visual feedbacks to heritage scientists, helping also to identify anomalous pixels in the images, which may present interesting features

Speaker: Serena Barone

BBB2026_Barone.pdf

3 AI-driven classification and pair-matching of 3D fragments of Nuragic pottery

This research focuses on the application of advanced digital technologies to the study and restoration of Nuragic pottery. The reconstruction of fragmented archaeological artefacts is a key challenge in cultural heritage research, enabling both the physical restoration of objects and a deeper understanding of their intact shape, typology, and function. Traditional manual reassembly requires time-consuming manual efforts by restorers and archaeologists, while recent advances in 3D digitisation and AI may provide new tools to face this challenge. A recent collaboration between the INFN-CHNet, the Soprintendenza Abap per la città metropolitana di Cagliari e le province di Oristano e Sud Sardegna and Fondazione Barumini has led to the development of an innovative digital pipeline. The project is structured around three main objectives:

1) Dataset: creation and publication of a large open-access dataset of 3D models of intact pottery vessels and their digitally fragmented version usable to train neural networks in object reconstruction tasks.

2) Classification: design, test and validation of a Deep Learning model capable of automatically identify pottery typologies processing shards in the form of 3D point clouds.

3) Partial reassembly: development of a Deep Learning model capable of evaluating adjacency matches between fragment pairs from the analysis of their geometry and fracture surfaces.

The State-Of-The-Art AI-based technologies for the analysis of 3D data will be employed (e.g. PointNet++, DGCNN, GATv2, PointTransformer etc.) with the goal to support the investigation and physical restoration of Nuragic pottery (1700 B.C.E-1000 B.C.E) found at sites under archaeological investigation by the Soprintendenza. Preliminary results of this work will be presented.

Speaker: Marta Magalini (Istituto Nazionale di Fisica Nucleare)

BBB2026_Magalini.pdf

Beams

Development of new technologies exploiting beams at different energies

Conveners: Alessandro Re (Istituto Nazionale di Fisica Nucleare), Massimiliano Clemenza (Istituto Nazionale di Fisica Nucleare)

4 On-site and laboratory X-ray scanning of cultural heritage: new challenges and opportunities

The demand from owners and institutions for radiographic imaging of artworks has been steadily increasing in both volume and complexity over time and substantially continues to grow. At the same time, available technology, such as CT systems and components, with a particular focus on data processing, are also rapidly evolving.
A comparison with other European groups revealed that the average standard of CT laboratories is quite high. The level of hardware and software integration, scanning accuracy, software development, and the number of high-end components (such as nanofocus X-ray tubes and high-resolution detectors), as well as the number of employees, is generally very good. This implies a considerable effort on our side to remain up to date.
Thanks to funding from the PNRR and fruitful collaborations, the CHNET node in Bologna has had the opportunity to partially renew its hardware and explore new fields such as photon-counting detectors, although not without facing new challenges.
The development of a new XY translation stage and the integration of a new high-resolution detector into our medium-size CT system have made the need for a data processing station urgent. The increased detector size and resolution generate very large dataset from a single scan and, when combined with translation to perform a tile-scanning CT, this leads to severe data-handling overload. Although residual PNRR funds should still be available and suitable to purchase a data-processing workstation, this need has not yet been met.
Mobile systems have been and continue to be our flagship; however, they also present significant operational and management challenges. Their use requires significant commitment from operators, and the number of campaigns that can be carried out each year is limited. Moreover, the risk of component failure, system misalignment as well as issues during the scanning process is not negligible, making mobile systems subject to a certain percentage of failures.
Nevertheless, over the past year we have carried out a significant number of analyses both on site and in our labs, and we present here a short overview upon them.
Along with intensive routine activity both in the laboratory and in situ, working on new standard and cutting-edge systems is certainly challenging. The collaboration with colleagues in high-energy physics on a project to develop a tomographic system based on a photon counting detector is about to conclude and, although this approach is theoretically very promising, substantial work remains to be done.
In light of all these considerations, we aim to define what our place is - or could be - within the broader landscape of cultural heritage analysis and what our contribution is or could be in view of the new evidence gathered this year.

Speaker: Dr Matteo Bettuzzi (Dipartimento di Fisica e Astronomia - Università di Bologna)

BBB2026_Bettuzzi.pdf

5 Neutrons meet Cultural Heritage: insights from Milano-Bicocca section

The INFN CHNet Milano-Bicocca group plays a pivotal role in advancing neutron-based methodologies for Heritage Science, focussing on the development of strictly non-destructive investigative protocols. A key research frontier currently being explored, in close synergy with the ISIS Neutron and Muon Source (UK), is the implementation of 4D neutron imaging. By exploiting resonance features at epithermal energies, this
approach extends the capabilities of standard neutron imaging to the study of complex, heterogeneous materials that remain opaque to conventional probes, while simultaneously providing isotopic sensitivity [1,2].

Under the framework of the CHNet_BRONZE project, the Milano-Bicocca section contributes to the optimization of three neutron-based techniques for the quantitative characterization of archaeological copper-based alloys. Specifically, our research leads the refining of Neutron Resonance Transmission Imaging (NRTI) to map and quantify alloy compositions, major components, and trace elements within the bulk of artifacts. The enhanced contrast capability of NRTI allows for a complementary understanding of ancient technological choices, manufacturing processes, and material heterogeneity.

Beyond methodological innovation, the Milano-Bicocca section is deeply integrated into national and international collaborations, employing a multi-modal approach which comprises neutron imaging and neutron diffraction. Recent successful applications include the investigation of meteorite collections in cross-collaboration with Pavia section, the analysis of the medieval Chiaravalle Cross’s golden filigree [3] in partnership with the ANSTO reactor, and the archaeometallurgical study of Nuragic-period objects [4], demonstrating the versatility and impact of neutron probes in cultural heritage preservation.

References
[1] Marcucci, G., et al. Mapping the elemental distribution in archaeological findings through advanced Neutron Resonance Transmission Imaging. Eur. Phys. J. Plus 139, 475 (2024). https://doi.org/10.1140/epjp/s13360-024-05222-y
[2] Marcucci, G., et al. Isotopic imaging with epithermal neutrons at the ISIS Neutron and Muon Source. Sci Rep 15, 19344 (2025). https://doi.org/10.1038/s41598-025-04283-y7
[3] Vigorelli, L., et al. High resolution neutron tomography as non-invasive tool for the study of a filigree from the Medieval Chiaravalle Cross. Archaeol Anthropol Sci 17, 138 (2025). https://doi.org/10.1007/s12520-025-02254-y
[4] Cataldo, M., et al. Compositional and morphological study of a Nuragic bronze figurine with neutron diffraction and neutron tomography, submitted to JINST.

Speaker: Dr Giulia Marcucci (Milano-Bicocca University and Istituto Nazionale di Fisica Nucleare, Milano-Bicocca section)

BBB2026_Marcucci.pdf

6 Muons to the rescue: activities at the Milano-Bicocca section

Based on the interaction of negative muons with matter and the following emission of X-ray radiation (specific to the atom which absorbed the muon), Muonic atom X-ray Emission Spectroscopy (µXES) is a very powerful method for elemental characterisation. Compared to common methods, which employ X-ray and electron beams, a significant advantage of using negative muons is their remarkable penetration depth. Moreover, thanks to their high energy (0.01-6 MeV), muonic X-rays are emitted from the bulk of the samples without significant photon self-absorption, thus enabling depth profiling. These characteristics, with the addition of a wide multi-elemental range (from Lithium to Uranium) and no residual activity left in the sample after irradiation, make µXES a unique tool for elemental analysis, with many different research projects published in the last few years [1, 2].
Over the last decade, the CHNet group at the INFN Milano-Bicocca section has actively collaborated with the ISIS Neutron and Muon Source to develop and apply μXES for Heritage Science studies. With two CSNV-founded projects and two PhD projects, the Milano-Bicocca section has built a unique knowledge and experience about negative muon spectroscopy. Within this framework, the CHNet-MAXI project, which concluded at the end of last year, focused on the development and validation of a non-destructive approach for the determination of lead isotopic ratios, a key tool for provenance studies of archaeological and historical artefacts. The project investigated isotopic shifts in high-energy muonic X-ray lines and associated γ-ray emissions following muon nuclear capture, through systematic measurements on certified Pb standards irradiated with negative muons, along with the development of a new digital acquisition system.
Besides development projects, the group continues the activities in support of users, including an ongoing collaboration with the Opificio delle Pietre Dure for the study of gilded surfaces and enamels. In the past years, experimental campaigns have been carried out at the ISIS facility to further contribute to the consolidation of the technique and the strengthening of the collaboration. In this context, Monte Carlo simulation software is routinely employed to support data interpretation. Simulation tools are used to optimise experimental configurations, reproduce experimental setup and correlate measured spectral features with sample composition and depth, highlighting the role of digital tools in the advancement of non-invasive cultural heritage analysis [3].

References
[1] Hillier, A.D., Blundell, S.J., McKenzie, I. et al. Nat Rev Methods Primers 2, 4, 2022.
[2] K. Ninomiya, Journal of Nuclear and Radiochemical Sciences 19, 2019, 8-13.
[3] M. Cataldo, A.D. Hillier et Al, Spectrochimica Acta Part B: Atomic Spectroscopy, 2025

Speaker: Matteo Cataldo (Istituto Nazionale di Fisica Nucleare)

BBB2026_Cataldo.pdf

7 INFN CHNet_BRONZE project: calibration of neutron techniques for the characterisation of archaeological bronzes

The study of ancient metallurgy, particularly the analysis of bronze and copper alloys, plays a crucial role in understanding the technological and artistic advancements of past civilisations. While traditional analysis involves invasive methods, such as SEM observation of cross sections and metallography, neutron-based techniques offer a highly promising solution for the non-invasive characterisation of the elemental, isotopic, and phase composition of the objects under study [1][2][3].
Here, the INFN CHNet_BRONZE project and the main results obtained so far are presented. The aim of the project is to develop and quantitatively calibrate techniques that exploit neutron probes for the analysis of Cu-based archaeological artefacts. The project assesses three bulk analysis methods: Time-of-Flight Neutron Diffraction (ToF-ND) that returns the phase composition of an artefact and the microstructure of metal phases; Neutron Resonance Transmission Imaging (NRTI) that allows bidimensional maps of the elemental and isotopic distribution; Bragg-Edge Neutron Transmission analysis (BENT) that explores the spatial distribution of metal phases, microstructure and texture of the artefact.
A set of purposely prepared copper-based reference samples, in powder and cast form with known composition and structure, are first used to define experimental and analytical methodologies. The results obtained with the three techniques allowed to derive calibration curves that can be later used for the interpretation of data obtained from historical bronze artefacts. This can help to answer questions of historical and cultural interest on archaeological bronzes, allowing to extract compositional and microstructure information from the bulk of the objects non-invasively.

REFERENCES
[1]. A. Fedrigo et al., Archaeol Anthropol Sci 10, 1249-1263 (2018)
[2]. A. De Palmas et al., Archaeol Anthropol Sci 13, 101 (2021)
[3]. G. Marcucci et al., European Physical Journal Plus, 139 (2024) 475.

Speaker: Miriana Marabotto (Istituto Nazionale di Fisica Nucleare)

BBB2026_Marabotto.pdf

17:10 Coffee break

Beauty

Diagnostic techniques for analysis of artworks and archeological assets

Convener: Valerio Graziani (Istituto Nazionale di Fisica Nucleare)

8 Development of an Automated 3D Scanning System for Multimodal Analysis of Cultural Heritage

The study of artworks, due to their material complexity and state of conservation, requires a multi-analytical approach based on complementary, non-invasive, and non-destructive techniques capable of providing compositional and morphological information. However, artworks are not always planar like paintings and may present complex geometries, such as sculptures or paintings on niches. It is therefore essential that the analytical systems used for artwork investigation cab be adapted to the surface’s geometry under examination.
In this context, we present a three-dimensional scanning system developed for the in-situ analysis of artworks. The system consists of a transportable XYZ scanning platform based on a collaborative six-axis robotic arm. The robotic arm has a maximum payload of 12.5 kg, a reach of 1300 mm, and a pose repeatability of ±0.05 mm, enabling accurate scanning of large and heterogeneous surfaces.
The platform is equipped with interchangeable sensor interfaces for laser profilometry, infrared reflectography, and macro-X-ray fluorescence (MA-XRF), allowing fully automated multimodal measurements. In addition, the set-up integrates a laser pointing system and a triple time-of-flight distance detection system, which enables the scanning head to adapt to the analysed surface, maintaining a constant safety distance and ensuring that the sensor remains perfectly parallel to the surface.
Dedicated Python-based software tools were developed to synchronize the robot motion with sensor acquisition, optimize scanning trajectories, and manage the large datasets generated during high-resolution automated measurements.

Speaker: Giulia Iorio (INFN Roma Tre, Università degli Studi Roma Tre)

BBB2026_Iorio_.pdf

9 POMPEII ADDICTION: Integrated Isotopic and Archaeometric Approaches to Life, Economy, and Mobility

This work presents an integrated archaeometric research program aimed at reconstructing subsistence strategies, economic networks, and population dynamics at Pompeii and in the Vesuvian area at the time of the AD 79 eruption. The project combines stable isotope analyses (δ¹³C, δ¹⁵N) on human, animal and botanical remains to refine dietary reconstructions and investigate agricultural and husbandry practices within the broader environmental and socio-economic framework. Baseline isotopic models are being developed to contextualize trophic relationships and resource availability, enabling a more robust interpretation of human paleodiet and economic organization.
Preliminary results indicate distinct dietary patterns among Vesuvian communities, with clear differentiation between terrestrial and marine resource use. The botanical and faunal isotopic baselines reveal diversified agricultural regimes and livestock management strategies, while radiocarbon analyses refine the chronology of occupation and post-eruptive reorganization.
Early dentin δ¹³C and δ¹⁵N data suggest structured and relatively homogeneous weaning practices, consistent with different patterns observed in other Roman sites such as Ostia Antica.
Strontium results demonstrate both local individuals and non-local signatures, supporting a scenario of moderate mobility and integration within wider economic networks.
Furthermore, an ongoing detailed multidisciplinary study of the site of Oplontis combines dietary reconstruction, mobility assessment, anthropological and paleopathological analysis to produce a comprehensive biocultural profile of its inhabitants.
Finally, the project includes a dissemination initiative which develops public-facing multimedia content for the exhibition “L’Altra Pompei - Vite comuni all’ombra del Vesuvio”, highlighting the analytical potential of organic materials in archaeological research.
Together, these integrated approaches advance our understanding of daily life, health, mobility, and economic organization in the Vesuvian region prior to the eruption, demonstrating the interpretive power of combined isotopic, chronological and anthropological analyses.

Speaker: Carmine Lubritto

10 The PITCH project: a laboratory-based phase contrast X-ray imaging system for Cultural Heritage analysis

Over the last few decades, X-ray imaging has improved significantly as a high-resolution, non-destructive, and non-invasive tool for radiography and tomography, solving a major challenge of detecting internal structures of samples in different applications. Furthermore, since the enhancement of synchrotron sources, extremely coherent radiation has been used for non-traditional contrast techniques [1]. The conventional absorption contrast method is particularly beneficial for characterising medium-density samples or distinguishing materials with different attenuation powers, but not for low-Z materials, which exhibit poor contrast under X-ray irradiation. Phase contrast X-ray imaging (PC) is instead an effective technique for detecting low contrast details in weakly absorbing samples. This method is based on the observation of the interference pattern between diffracted and non-diffracted waves, caused by spatial variations in the real component of the refractive index introduced by a sample placed in the wave path [2]. PC X-ray imaging has great potential in medical, material science, and cultural heritage (CH) applications since it can deliver high-quality information on microstructural characteristics. A coherent X-ray beam is an appropriate instrument for developing phase-sensitive X-ray imaging; as a result, this technology has already been deployed at synchrotron facilities [3]. Among the PC methodologies, X-ray Grating Interferometry (GI), which provides differential phase and scattering (dark-field) images in addition to the standard absorption-based image, enables the realisation of systems that can be implemented in a laboratory setting. This framework opens interesting opportunities in the CH field, including conservation, archaeology, and anthropology.
The PITCH project (funded in the framework of PRIN2022) aims at designing, developing and characterising a laboratory GI-PC imaging setup based on a liquid-anode X-ray source, as well as at investigating and developing different data acquisition methods and algorithms for signal extraction and tomographic reconstruction. The system will be optimised for the analysis of various CH materials to characterise and better understand specific related issues.
References
[1] R. Fitzgerald, “Phase‐Sensitive X‐Ray Imaging,” Phys Today, vol. 53, no. 7, pp. 23–26, 2000
[2] M. Endrizzi, “X-ray phase-contrast imaging” Elsevier B.V., 2018
[3] R. Fitzgerald et al., “Phase objects in synchrotron radiation hard X-ray imaging,” Cambridge Univ Press, 2009

Speaker: Alessandro Re (University of Torino and Istituto Nazionale di Fisica Nucleare)

BBB2026_Re.pdf

11 Integrating Analytical Techniques to Extend the Spectral Range from X-rays to Infrared: A Multi-Scale Approach for Painting Materials

The characterization of Cultural Heritage (CH) materials requires a multiscale and multi-analytical approach to fully understand the complexity of both original constituents and degradation products. This work discusses the potential of integrating diverse spectroscopic techniques across different spatial resolutions (micro and macro scale). Two primary technological approaches are highlighted:
At the macroscopic scale, the implementation of an in situ multisensor scanning system. This methodology enables the simultaneous or sequential acquisition of elemental and molecular data, specifically through Macro X-Ray Fluorescence (MA-XRF), UV-induced Fluorescence (UVF), and external reflection FT-IR (rFTIR), directly on the artwork’s surface. The system is engineered to generate extensive hyperspectral chemical maps, allowing for the correlation of diverse material signals while ensuring the total physical integrity of the object.
At the microscopic scale, the development of an integrated workflow for high-resolution imaging on painted surfaces. This approach involves a Vis-UV microscope coupled with a hyperspectral camera and micro-FTIR spectroscopy to define molecular composition, complemented by SEM-EDX analysis to achieve precise elemental characterization. All of these techniques can be performed both on the sample surface and across stratigraphic cross-sections.
By combining these two methodologies, it is possible to obtain a global understanding of the artwork's surface distribution to a precise chemical characterization of its microscopic layers, providing a complete view of the material history, artistic techniques, and conservation state.

Speaker: Lucilla Pronti (INFN-LNF)

BBB2026_Pronti.pdf

20:30 Social dinner

Ristorante la Vecchia Pineta
P.le dell'Aquilone, 4, 00122 Lido di Ostia RM

Friday 6 March

News from CHNet working groups

11:00 Coffee break

CHNet roundtable

Convener: Francesco Taccetti (Istituto Nazionale di Fisica Nucleare)

11:30 Coffee break

Closing remarks

Conveners: Francesco Taccetti (Istituto Nazionale di Fisica Nucleare), Prof. Luca Tortora (INFN Roma Tre)