Newer
Older
\documentclass[a4paper]{jpconf}
\usepackage[T1]{fontenc}
\usepackage[utf8]{inputenc}
\usepackage{graphicx}
\usepackage{url}
\begin{document}
\title{DHLab: a digital library for the INFN Cultural Heritage Network}
\author{F. Proietti$^1$, L. dell'Agnello$^1$, F. Giacomini$^1$}
\address{$^1$ INFN-CNAF, Bologna, IT}
\ead{fabio.proietti@cnaf.infn.it}
\begin{abstract}
DHLab, as part of the Cultural Heritage Network (CHNet) promoted by
INFN, is a cloud-based environment to process, visualise and analyse
data acquired from members of the network and that will be provided
to technical and non-technical users. DHLab is under development and
currently its main features are a cloud service to upload and manage
the data, a form to assign metadata to uploaded datasets and a
service used to analyze data obtained from XRF measurements.
\end{abstract}
\section{Introduction}
CHNet\footnote{http://chnet.infn.it/} is a network composed by several
INFN teams who devote their expertise in physics research to the study
and diagnostics of Cultural Heritage. By using their existing instruments,
developed for Nuclear Physics, or even by building new ones,
INFN laboratories started to address the needs of archaeologists,
historians, art historians, restorers and conservators. This unified
knowledge can provide useful indications about the correct procedures
to be applied for restoration or conservation, could be important to
date or verify, for example, the authenticity of an artwork or study
the provenance of raw material in order to retrace ancient trade
routes. In this context the purpose of the DHLab is to host all the
data acquired by the CHNet laboratories, together with the
descriptions and annotations added by humanists.
\section{Architecture}
The infrastructure system, shown in figure~\ref{fig:architecture},
follows a cloud-based model and can be divided in multiple modular
frontends, providing the interface towards the clients, and a
monolithic backend service.
\begin{figure}[ht]
\begin{center}
\caption{\label{fig:architecture}High level overview of DHLab
architecture}
\end{center}
\end{figure}
The frontend includes three main blocks: a cloud service, a metadata
form and an application service. Of these, the metadata form, used to
fill details about a work or an analysis (see
section~\ref{sec:metadata-form}), is usable also while being offline;
the requirement addresses the use case of an operator who, while
disconnected from the network, needs to fill the metadata saving them
as a file on the local machine. The same requirement may be at least
partly satisfied also for the application services.
On the backend side, which is only partially implemented and not yet
even fully designed, we currently expect to have a listener, to
dispatch client requests, two data stores, one for user profiles and
the other for actual datasets, and a set of auxiliary services, for
example to automate the filling of the metadata form and to
standardize some of its fields (see again
section~\ref{sec:metadata-form}).
The entire system is hosted at the CNAF data center.
\section{Technologies and protocols}
As stated above, the design of the system is not yet complete and we
are still investigating different options to address the challenges we
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
Open aspects concern:
\begin{itemize}
\item the data model, which must accomodate both datasets (possibly
composed of multiple files), the corresponding metadata and a
mechanism to link them together;
\item the authentication and authorization model, which should use as
much as possible standard web technologies and have flexible
mechanisms to authenticate users coming from different institutions,
leveraging their own Identity Providers;
\item how to access the available storage from a client, both to
upload datasets and their metadata and subsequently access them.
\end{itemize}
The experimentation makes use of an installation of
NextCloud~\cite{ref:nextcloud}, an open-source suite of client-server
software for creating and using file hosting services, with
functionality often extended through the use of plugins.
Authentication is based on OpenID Connect~\cite{ref:oidc} and makes
use of the INDIGO-IAM~\cite{ref:iam} service, an Identity and Access
Management product developed within the EU-funded
INDIGO-DataCloud~\cite{ref:indigo} project. INDIGO-IAM offers a
service to manage identities, user enrollment, group membership,
attributes and policies to access distributed resources and services
in a homogeneous and interoperable way; hence it represents a perfect
match to manage users, groups and resources of the CHNet
organization. In particular INDIGO-IAM delegates the authentication of
a user to their home institution identity provider under a trust
agreement.
NextCloud offers also the possibility to access data via the WebDAV
protocol, allowing users to mount the remote storage on their local
machine and see it as if it were a local disk. This feature becomes
useful when interaction through a web browser is not the most
effective tool, for example for batch or bulk operations.
\section{Metadata Form}
\label{sec:metadata-form}
The Metadata form is a web application whose purpose is to associate
metadata with art works, measurement campaigns and analysis
results. The application, written in Typescript~\cite{ref:typescript}
and based on the Angular 2 framework~\cite{ref:angular2}, is under
development; the main deployment option foresees its integration into
the cloud platform, but the combination with
Electron~\cite{ref:electron} makes a desktop application a viable
alternative.
As shown in figure~\ref{fig:metadataSchema}, to fill the metadata form
a user can follow two paths: they can create a \textit{campaign} and
associate it with multiple \textit{sessions} and \textit{analyses} or
they can store information about a single \textit{analysis}. In
particular, each \textit{analysis} can be associated with one or more
\textit{datasets}, the studied \textit{object} (i.e.,~an art work) and
all the information about its \textit{type}, \textit{author},
\textit{holder}, \textit{owner}, etc. In addition, users can provide
information about the analysis type, the operator who performed the
analysis, the devices, components and software used to scan, create or
read the resulting dataset. When completed, the resulting form,
translated into a JSON file, can be saved locally or uploaded to the
remote storage.
\begin{figure}[ht]
\begin{center}
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
\end{center}
\caption{\label{fig:metadataSchema}Schema of the sections included
in the metadata description.}
\end{figure}
\section{Application services}
DHLab is also designed to provide visualization and analysis services
for some of the stored datasets. Currently a proof-of-concept
application is available, to visualize and perform some analysis of
images obtained from XRF scans~\cite{ref:xrf}.
\section{Conclusions}
DHLab is a project born from the need to group, share, catalogue and
reuse data that comes from measurements and analyses of cultural
heritage works. It aims at being flexible and usable by persons
covering different roles: physicists, computer scientists, cultural
heritage operators. The system is designed and deployed around a core
Cloud-based infrastructure, but some of its parts must be functioning
in offline situations.
A web application for filling a form with metadata to be associated to
collected datasets according to an agreed-upon schema is being
developed.
Other web applications are foreseen for the visualization and analysis
of the stored datasets, starting from those coming from XRF,
radiocarbon and thermoluminescence analysis.
\section*{References}
\begin{thebibliography}{9}
\bibitem{ref:nextcloud} NextCloud \url{https://nextcloud.com/}
\bibitem{ref:oidc} OpenId Connect \url{https://openid.net/connect}
\bibitem{ref:iam} A Ceccanti, E Vianello, M Caberletti. (2018,
May). INDIGO Identity and Access Management
(IAM). Zenodo. \url{http://doi.org/10.5281/zenodo.1874790}
\bibitem{ref:indigo} The INDIGO-DataCloud project
\url{https://www.indigo-datacloud.eu/}
\bibitem{ref:typescript} Typescript language
\url{https://www.typescriptlang.org/}
\bibitem{ref:angular2} Angular 2 framework
\url{https://angular.io/}
\bibitem{ref:electron} Electron
\url{https://electronjs.org/}
\bibitem{ref:xrf} Cappelli L, Giacomini F, Taccetti F, Castelli L,
dell'Agnello L. 2016. A web application to analyse XRF scanning data. INFN-CNAF
Annual Report. \url{https://www.cnaf.infn.it/annual-report}
\end{thebibliography}
\end{document}