Detecting social patterns within 20th century documentary photos: a deep learning based approach

The job of a historian is to understand what happened in the past, resorting in many cases to written documents as a firsthand source of information. Text, however, does not amount to the only source of knowledge. Pictorial representations, in fact, have also accompanied the main events of the historical timeline. In particular, the opportunity of visually representing circumstances has bloomed since the invention of photography, with the possibility of capturing in real-time the occurrence of a specific events. Thanks to the widespread use of digital technologies (e.g. smartphones and digital cameras), networking capabilities and consequent availability of multimedia content, the academic and industrial research communities have developed artificial intelligence (AI) paradigms with the aim of inferring, transferring and creating new layers of information from images, videos, etc. Now, while AI communities are devoting much of their attention to analyze digital images, from an historical research standpoint more interesting results may be obtained analyzing analog images representing the pre-digital era. Within the aforementioned scenario, the aim of this work is to analyze a collection of analog documentary photographs, building upon state-of-the-art deep learning techniques. In particular, the analysis carried out in this thesis aims at producing two following results: (a) produce the date of an image, and, (b) recognizing its background socio-cultural context,as defined by a group of historical-sociological researchers. Given these premises, the contribution of this work amounts to: (i) the introduction of an historical dataset including images of “Family Album” among all the twentieth century, (ii) the introduction of a new classification task regarding the identification of the socio-cultural context of an image, (iii) the exploitation of different deep learning architectures to perform the image dating and the image socio-cultural context classification.

Industrial accidents triggered by lightning: causes and consequences

Natural hazards affecting industrial installations could directly or indirectly cause an accident or series of accidents with serious consequences for the environment and for human health. Accidents initiated by a natural hazard or disaster which result in the release of hazardous materials are commonly referred to as Natech (Natural Hazard Triggering a Technological Disaster) accidents. The conditions brought about by these kinds of events are particularly problematic, the presence of the natural event increases the probability of exposition and causes consequences more serious than standard technological accidents. Despite a growing body of research and more stringent regulations for the design and operation of industrial activities, Natech accidents remain a threat. This is partly due to the absence of data and dedicated risk-assessment methodologies and tools. Even the Seveso Directives for the control of risks due to major accident hazards do not include any specific impositions regarding the management of Natech risks in the process industries. Among the few available tools there is the European Standard EN 62305, which addresses generic industrial sites, requiring to take into account the possibility of lightning and to select the appropriate protection measures. Since it is intended for generic industrial installations, this tool set the requirements for the design, the construction and the modification of structures, and is thus mainly oriented towards conventional civil building. A first purpose of this project is to study the effects and the consequences on industrial sites of lightning, which is the most common adverse natural phenomenon in Europe. Lightning is the cause of several industrial accidents initiated by natural causes. The industrial sectors most susceptible to accidents triggered by lightning is the petrochemical one, due to the presence of atmospheric tanks (especially floating roof tanks) containing flammable vapors which could be easily ignited by a lightning strike or by lightning secondary effects (as electrostatic and electromagnetic pulses or ground currents). A second purpose of this work is to implement the procedure proposed by the European Standard on a specific kind of industrial plant, i.e. on a chemical factory, in order to highlight the critical aspects of this implementation. A case-study plant handling flammable liquids was selected. The application of the European Standard allowed to estimate the incidence of lightning activity on the total value of the default release frequency suggested by guidelines for atmospheric storage tanks. Though it has become evident that the European Standard does not introduce any parameters explicitly pointing out the amount of dangerous substances which could be ignited or released. Furthermore the parameters that are proposed to describe the characteristics of the structures potentially subjected to lightning strikes are insufficient to take into account the specific features of different chemical equipment commonly present in chemical plants.