Andrea Lingua

Monocular depth estimation (MDE) has reached notable maturity in computer vision, yet its application to UAV-based architectural heritage documentation remains underexplored. This study assesses whether the depth foundation model Depth Anything V2 can be transferred from terrestrial to aerial imagery. The analysis relies on MDE4BH, a benchmark of over 3,000 UAV images covering ten heterogeneous heritage scenarios (urban areas, façades, towers, villas, domes, and archaeological sites). Masked photogrammetric depth maps serve as metric reference for calibration, validation, and supervised retraining. Two baseline configurations are evaluated: a relative model with scene-specific linear rescaling and the direct application of the metric model. The rescaled relative model shows acceptable performance in several subsets, whereas the metric model exhibits systematic bias, weak consistency, and scale collapse due to domain shift between terrestrial training data and aerial acquisition geometry. To address these limitations, a two-step fine-tuning strategy is introduced, focusing on the decoder and regression head. The first stage uses mainly oblique UAV images; the second integrates oblique and nadir views to improve viewpoint generalization. The adapted model significantly reduces bias and enhances metric stability across the benchmark. However, residual errors remain spatially structured, with clustering and recurrent artefacts near object boundaries, multi-level roofs, and radiometrically heterogeneous surfaces. Although accuracy is still insufficient for demanding metric applications, the results support the use of MDE as a complementary source for thematic interpretation, scene understanding, robotics, navigation, and related tasks where strict geometric precision is not required.

The POSEIDON project aims to develop scalable and repeatable approaches for monitoring Posidonia Oceanica (PO) meadows, a key Mediterranean habitat that supports coastal ecosystem services and long-term blue-carbon storage, yet they are increasingly threatened by warming waters and cumulative human pressures. This work presents a satellite-based workflow for benthic habitat mapping that combines Sentinel-2 multispectral imagery, ancillary bathymetry, and deep-learning semantic segmentation. Sentinel-2 Level-2A data and bathymetry were integrated into multi-band inputs on a common 10 m grid, with analysis restricted to water pixels. A wall-to-wall reference map was generated by harmonising existing habitat products into six benthic classes for supervised model training and evaluation. U-Net and DeepLabv3 architectures with a ResNet backbone were tested for a representative September 2015 scene. The workflow was first assessed in the Culuccia peninsula, where it achieved an overall accuracy of 0.830 and a Kappa coefficient of 0.786. It was then successfully transferred to the Capo Testa - Punta Falcone Marine Protected Area (MPA), where the best-performing configuration reached an overall accuracy of 0.882 and a Kappa coefficient of 0.843. These results show that open-access satellite data combined with robust semantic segmentation models can provide a reliable and non-destructive framework for seagrass mapping in complex coastal environments.

Historical masonry hydraulic infrastructures are affected by complex degradation processes, including vegetation growth, moisture-related anomalies, and salt efflorescence, whose detection requires non-invasive, repeatable, and scalable diagnostic approaches. This study proposes a multi-scale workflow for detecting and classifying degradation phenomena affecting the Cavour Canal water bridge, a nineteenth-century masonry structure in northern Italy. The methodology combines UAV-based multispectral orthophotos and close-range hyperspectral imagery within a common Object-Based Image Analysis (OBIA) framework. The multispectral workflow was designed for façade-scale screening, whereas the hyperspectral workflow was used to support the detailed characterisation of selected sectors through detailed spectral characterisation. Multiple supervised classifiers, including Support Vector Machine (SVM), k-Nearest Neighbours (kNN), Decision Tree (DT), Random Trees (RT), and Naïve Bayes (NB), were tested on both datasets. The results show that the multispectral workflow is effective for identifying vegetation and broad water-related anomalies, with kNN providing the best overall performance, while the hyperspectral workflow improves the discrimination of subtle surface alterations, particularly efflorescence, with SVM yielding the most stable results across the tested configurations. Overall, the proposed methodology demonstrates the value of a hierarchical multispectral and hyperspectral workflow for non-invasive degradation mapping of historical masonry hydraulic infrastructures.

The accurate digital documentation of Cultural Heritage (CH) assets demands workflows capable of integrating heterogeneous, multiscale datasets while preserving both geometric fidelity and radiometric completeness. This paper presents the first results of the AI-based processing pipeline developed within the HERITALISE project (Horizon Europe, 2025–2028), applied to three multiscale case studies at the Reggia di Venaria Reale (Turin, Italy): an outdoor-indoor UAV photogrammetric survey, a kinematic SLAM acquisition of a contemporary sculpture garden, and a close-range dataset of an 18th-century decorative artefact. 3D Gaussian Splatting (3DGS) is evaluated as a novel view synthesis method across all three scenarios, demonstrating strong photorealistic rendering capabilities, particularly for complex material properties and geometrically challenging interiors, whilst highlighting current limitations for metric surveying applications. A two-stage crack detection workflow, combining tile-based text-prompted segmentation with SAM3 and multiview ray-based reprojection onto the reconstructed mesh, is validated on UAV imagery, achieving an 84.9% ray–mesh intersection rate. Finally, a standardised evaluation framework is proposed, encompassing adaptive, scale-dependent geometric and radiometric metrics organised into reference-based and no-reference assessment scenarios, aggregated into a transparent synthetic quality score with three adaptive quality classes. The proposed methodology contributes toward a reproducible, sensor-agnostic standard for the assessment of AI-generated CH documentation products.

Posidonia oceanica banquettes represent key coastal features in the Mediterranean, providing ecosystem services such as shoreline protection, sediment retention, and carbon storage. Despite their ecological importance, their spatial and temporal dynamics remain poorly quantified due to the lack of standardized monitoring approaches. This study presents a UAV-based photogrammetric methodology for the multitemporal analysis of banquette morpho dynamics, integrating NRTK-supported surveys, Structure-from-Motion processing, and uncertainty-aware DEM differencing. The method was applied to Maurizio Vitale Beach (Culuccia Island, Sardinia, Italy), a protected and low-impact coastal environment, using four UAV surveys acquired between 2022 and 2025. High-resolution Digital Elevation Models (DEMs) were generated and compared to quantify volumetric changes across seasonal and interannual timescales. Results show a highly dynamic system, characterized by alternating phases of erosion, redistribution, and accumulation. Seasonal analysis (June–October 2024) revealed moderate volumetric losses (−41.3 m³), mainly affecting the seaward edge, with localized deposition in inner sectors. Interannual comparison (2022–2024) highlighted a more pronounced net loss (−74.2 m³). Conversely, the 2024–2025 analysis showed significant recovery, with widespread accretion (+95.8 m³), indicating strong temporal variability. The proposed approach proved effective in capturing both horizontal and vertical changes, distinguishing between boundary retreat and surface lowering processes. These results demonstrate the potential of UAV photogrammetry as a reliable and scalable tool for standardized banquette monitoring, supporting coastal management strategies and contributing to the objectives of the POSEIDON project.

Vegetation mapping in alpine environments is essential for monitoring ecosystem dynamics and climate change impacts, yet remains challenging when using very high-resolution UAV imagery under limited labeled data. This study proposes a data-centric, pixel- based classification framework for class-level vegetation mapping using multispectral UAV data acquired in an alpine study area. The approach prioritizes improving data representation rather than increasing model complexity. To address label scarcity, a feature-rich dataset was constructed by integrating spectral information, vegetation indices, and lightweight spatial descriptors to enhance class separability. Classification was performed using XGBoost, which is well suited for multispectral tabular data and robust under imbalanced conditions. The results show consistent classification performance across vegetation types and demonstrate the effectiveness of dataset enrichment under limited supervision, highlighting the importance of feature representation in data-scarce scenarios.

Mountain ecosystems are highly sensitive to climate change and require spatially explicit monitoring tools to support adaptive management. Within the framework of the Interreg-ALCOTRA “ACLIMO” project, this study investigates land cover dynamics in the Gesso Valley (Maritime Alps, Italy) over the period 2010–2021 using deep learning–based classification of high-resolution aerial orthophotos integrated with climate data analysis. Multi-temporal RGB and NIR imagery (2010, 2018, 2021) was classified using convolutional neural networks (U-Net and MMSegmentation) in ArcGIS Pro, with CORINE Land Cover datasets used for training. The best-performing model, based on CLC + Backbone 2018, achieved an overall accuracy of 82%, increasing to 87% after fine-tuning. Change detection revealed a general shift towards increased vegetation cover, while climate analysis based on regional weather stations (1990–2021) identified a warming trend of +0.4 °C/decade and recent drier conditions. Logistic regression highlighted significant associations between land cover transitions and climate anomalies, with temperature positively influencing change probability (OR = 1.40). The study demonstrates the potential of operational GIS-integrated deep learning workflows for climate change monitoring in complex alpine environments under real-world data constraints.

In the context of tunnel excavation with Tunnel Boring Machines (TBM), the possibility of quickly determining, on site and with low-cost tools, both the grain size curve and the volume of the excavated muck, can support the correct setting of the machine and the possible reuse of the material itself for applications such as structural concrete for tunnel linings, road embankments and fillings and so on. For this purpose, unconventional data acquisition methodologies were investigated, such as photogrammetry and LiDAR sensors mounted on smartphones, since they are time-effective from a processing point of view and can be easily adapted and modulated with relatively low budgets. In particular, for the determination of the grain size curve, methods already present in the state-of-the-art were tested, such as BASEGRAIN (expeditiously studied for river pebbles with rounded shapes and not with sharped edges like those of this study), and artificial intelligence techniques for automatic segmentation of rocks from images (Mask R-CNN). For the Volumetric estimation tests were conducted both on the entire sample and on individual grains in order to correctly estimate the volume of the whole excavated sample. The results showed how it is possible to obtain with these techniques on one hand a proper approximation of the granulometric curve, fully comparable with laboratory data, and on the other hand cumulative curves.