Geometric Features Analysis for the Classification of Cultural Heritage Point Clouds

Abstract

<p><strong>Abstract.</strong> In the last years, the application of artificial intelligence (Machine Learning and Deep Learning methods) for the classification of 3D point clouds has become an important task in modern 3D documentation and modelling applications. The identification of proper geometric and radiometric features becomes fundamental to classify 2D/3D data correctly. While many studies have been conducted in the geospatial field, the cultural heritage sector is still partly unexplored. In this paper we analyse the efficacy of the geometric covariance features as a support for the classification of Cultural Heritage point clouds. To analyse the impact of the different features calculated on spherical neighbourhoods at various radius sizes, we present results obtained on four different heritage case studies using different features configurations.</p>

A Hierarchical Machine Learning Approach for Multi-Level and Multi-Resolution 3D Point Cloud Classification

“…Starting from these considerations, the developed classification method uses a Random Forest (RF) classifier [54], following the successful supervised approach based on geometric features introduced in [34] and the study on features importance in Random Forest [20,33]. Compared to DL approaches, RF methods do not need a significant amount of manually annotated datasets to be effective.…”

“…More specifically to the heritage field, a supervised learning approach which transfers the classification information from 2D textures to 3D models is proposed in [21]. Grilli et al [33] presented a classification approach that works directly on point clouds, training a Random Forest (RF) classifier with geometric features. The method iteratively extracts the most relevant features considering a set of geometric characteristics strictly related to the architectural element dimensions.…”

“…For both case studies, the considered local neighbourhoods are directly related to the dimension of the architectural elements. Moreover, as various features affect the classification results in different ways, the geometric features with the smallest impact on the training process have been iteratively removed [33]. The experiments showed that, independently from the object to be classified and the point cloud resolution, the features with a more significant impact on the classification results are (Figure 8):…”

Networking Historic Environmental Standards to Address Modern Challenges for Sustainable Conservation in HBIM

“…With characteristic end-to-end architecture, which trains simultaneously representation learning and classification, the model allows the tailoring of feature learning to the needs of the classification task at hand. This approach overcomes the need of having input features generated independently of the training process, which is typical of traditional machine learning methods such as kNN, Support Vector Machines and classic Classification Trees [67][68][69]. Furthermore, a sensible sizing and training of the model and the very existence of a critical mass of well documented examples of historic buildings designed using standard architectural components, allowed the exploitation of deep learning effectively and overcame the risks of poor performance connected to overfitting of images.…”

Multi-Scalar Surveys for Complex Architectures

“…In this study, the elaborated point clouds, recorded through different acquisition phases, were also used to compare and analyse the metric data deriving from different technologies (Grilli et al, 2019). The comparison was carried out through a critical and direct analysis of the processed data, along with the use of specific software with which the metric data of the different point clouds were examined and compared (Bastonero et al, 2014).…”