A cloud-native application for digital restoration of Cultural Heritage using nuclear imaging: THESPIAN-XRF

Abstract

Artificial Intelligence for digital REStoration of Cultural Heritage (AIRES-CH) is a project focused on a creation of a cloud-native web application for the digital restoration of pictorial artworks through computer vision technologies applied to nuclear imaging raw data. In a previous work,it was shown that the task of associating an RGB colour image to a X-ray fluorescence (XRF) imaging raw data is feasible by means of a multidimensional neural network, and it was performed hyperparameter optimisation of the models. In this contribution we describe how the trained neural network is employed in the cloud-native web application for XRF raw data real time analysis. The RESTful API offering the Neural Network(s) has been developed using three frameworks and two languages, and a benchmark of its performances has been conducted. In the end, we comment the different outcomes of the benchmarking for the two different neural network branches. In the end, we comment the different outcomes of the benchmarking for the two different neural network branches.