Revolutionizing urban mapping: deep learning and data fusion strategies for accurate building footprint segmentation.

In the dynamic urban landscape, understanding the distribution of buildings is paramount. Extracting and delineating building footprints from high-resolution images, captured by aerial platforms or satellites, is essential but challenging to accomplish manually, due to the abundance of high-resolution data. Automation becomes imperative, yet it introduces complexities related to handling diverse data sources and the computational demands of advanced algorithms. The innovative solution proposed in this paper addresses some intricate challenges occurring when integrating deep learning and data fusion on Earth Observed imagery. By merging RGB orthophotos with Digital Surface Models, deriving from the same aerial high-resolution surveys, an integrated consistent four-band dataset is generated. This unified approach, focused on the extraction of height information through stereoscopy utilizing a singular source, facilitates enhanced pixel-to-pixel data fusion. Employing DeepLabv3 algorithms, a state-of-the-art semantic segmentation network for multi-scale context, pixel-based segmentation on the integrated dataset was performed, excelling in capturing intricate details, particularly when enhanced by the additional height information deriving from the Digital Surface Models acquired over urban landscapes. Evaluation over a 21 km2 area in Turin, Italy, featuring diverse building frameworks, showcases how the proposed approach leads towards superior accuracy levels and building boundary refinement. Notably, the methodology discussed in the present article, significantly reduces training time compared to conventional approaches like U-Net, overcoming inherent challenges in high-resolution data automation. By establishing the effectiveness of leveraging DeepLabv3 algorithms on an integrated dataset for precise building footprint segmentation, the present contribution holds promise for applications in 3D modelling, Change detection and urban planning. An approach favouring the application of deep learning strategies on integrated high-resolution datasets can then guide decision-making processes facilitating urban management tasks.

Microgeophysics and geomatics data integration reveals the internal fracturing conditions of the statue of Ramses II (Museo Egizio, Torino, Italy).

The combined acquisition of 3D ultrasonic tomography and radar scans is growing for cultural heritage diagnostics. Both methods proved to be efficient in the detection and location of fractures and weaknesses within the investigated artefacts. Although the two techniques are widely applied together, an integrated approach for data interpretation is still missing. We present the results of radar and ultrasonic prospections carried out on the statue of the young Ramses II, an absolute masterpiece of the Egyptian art preserved in the collection of the Museo Egizio of Torino (Italy). Geophysical results are incorporated within the 3D model of the statue retrieved from total station measurements, ground-based and handheld laser scanning. A data integration approach is then proposed for the joint interpretation of the geophysical results, exploiting the final ultrasonic velocity model and radar attribute analysis (i.e. local dissimilarity computation) to define a combined damage index. The proposed methodology is efficient in fracture detection and location and improves the readability of the final results also for non-expert geophysical interpreters, offering guidance to the museum for preservation and restoration of the masterpiece.

Inertial sensors for smartphones navigation.

The advent of smartphones and tablets, means that we can constantly get information on our current geographical location. These devices include not only GPS/GNSS chipsets but also mass-market inertial platforms that can be used to plan activities, share locations on social networks, and also to perform positioning in indoor and outdoor scenarios. This paper shows the performance of smartphones and their inertial sensors in terms of gaining information about the user's current geographical locatio n considering an indoor navigation scenario. Tests were carried out to determine the accuracy and precision obtainable with internal and external sensors. In terms of the attitude and drift estimation with an updating interval equal to 1 s, 2D accuracies of about 15 cm were obtained with the images. Residual benefits were also obtained, however, for large intervals, e.g. 2 and 5 s, where the accuracies decreased to 50 cm and 2.2 m, respectively.