Algorithms for on-line image registration from multiple views

Images of a scene, static or dynamic, are generally acquired at different epochs from different viewpoints. They potentially gather information about the whole scene and its relative motion with respect to the acquisition device. Data from different (in the spatial or temporal domain) visual sources can be fused together to provide a unique consistent representation of the whole scene, even recovering the third dimension, permitting a more complete understanding of the scene content. Moreover, the pose of the acquisition device can be achieved by estimating the relative motion parameters linking different views, thus providing localization information for automatic guidance purposes. Image registration is based on the use of pattern recognition techniques to match among corresponding parts of different views of the acquired scene. Depending on hypotheses or prior information about the sensor model, the motion model and/or the scene model, this information can be used to estimate global or local geometrical mapping functions between different images or different parts of them. These mapping functions contain relative motion parameters between the scene and the sensor(s) and can be used to integrate accordingly informations coming from the different sources to build a wider or even augmented representation of the scene. Accordingly, for their scene reconstruction and pose estimation capabilities, nowadays image registration techniques from multiple views are increasingly stirring up the interest of the scientific and industrial community. Depending on the applicative domain, accuracy, robustness, and computational payload of the algorithms represent important issues to be addressed and generally a trade-off among them has to be reached. Moreover, on-line performance is desirable in order to guarantee the direct interaction of the vision device with human actors or control systems. This thesis follows a general research approach to cope with these issues, almost independently from the scene content, under the constraint of rigid motions. This approach has been motivated by the portability to very different domains as a very desirable property to achieve. A general image registration approach suitable for on-line applications has been devised and assessed through two challenging case studies in different applicative domains. The first case study regards scene reconstruction through on-line mosaicing of optical microscopy cell images acquired with non automated equipment, while moving manually the microscope holder. By registering the images the field of view of the microscope can be widened, preserving the resolution while reconstructing the whole cell culture and permitting the microscopist to interactively explore the cell culture. In the second case study, the registration of terrestrial satellite images acquired by a camera integral with the satellite is utilized to estimate its three-dimensional orientation from visual data, for automatic guidance purposes. Critical aspects of these applications are emphasized and the choices adopted are motivated accordingly. Results are discussed in view of promising future developments.

On-line adaptive visual tracking

Visual tracking is the problem of estimating some variables related to a target given a video sequence depicting the target. Visual tracking is key to the automation of many tasks, such as visual surveillance, robot or vehicle autonomous navigation, automatic video indexing in multimedia databases. Despite many years of research, long term tracking in real world scenarios for generic targets is still unaccomplished. The main contribution of this thesis is the definition of effective algorithms that can foster a general solution to visual tracking by letting the tracker adapt to mutating working conditions. In particular, we propose to adapt two crucial components of visual trackers: the transition model and the appearance model. The less general but widespread case of tracking from a static camera is also considered and a novel change detection algorithm robust to sudden illumination changes is proposed. Based on this, a principled adaptive framework to model the interaction between Bayesian change detection and recursive Bayesian trackers is introduced. Finally, the problem of automatic tracker initialization is considered. In particular, a novel solution for categorization of 3D data is presented. The novel category recognition algorithm is based on a novel 3D descriptors that is shown to achieve state of the art performances in several applications of surface matching.

Tra società informazionale e prosumerismo: il citizen journalism e la partecipazione on line

I temi della ricerca riguardano il rapporto fra avvento del web e la modificazione dei processi di formazione di identità personale e sociale, della percezione dello spazio e del tempo, del prosumerismo digitale e delle varie forme di partecipazione ed associazione. Centrale è stata l’analisi del rapporto fra il Web 2.0 e la trasformazione delle forme di comunicazione a vari livelli, sia personali che sociali. Partendo da una analisi dei contesti socio-economici globali che hanno trasformato la società moderna nella società informazionale, è stato impostato un percorso di ricerca che approfondisse gli attuali criteri di strutturazione della propria identità, alla luce dell’avvento dei social network e delle reti virtuali di comunicazione come strumento preferenziale di socializzazione. La realtà delle reti sociali è stata analizzata in un’ottica di aggregazione spontanea mirata tanto alla comunicazione quanto alla tutela dei consumatori, e le trasformazioni portate dal Web 2.0 sono state la chiave di lettura per ridefinire i parametri della partecipazione dal basso generata dalla rete. Per comprendere la portata di tali trasformazioni nel contesto italiano è stato impostato un paragone tra l’uso del web negli Stati Uniti e in Italia, avendo le recente campagne elettorali dimostrato l’importanza del web nella partecipazione politica bottom-up; il percorso di ricerca ha dunque affrontato una comparazione di due casi, quello italiano e quello statunitense, finalizzato a comprendere l’attuale ruolo dell’utente nelle dinamiche di comunicazione mediatica. Per focalizzare al meglio le trasformazioni sociali generate dalla partecipazione on line è stato infine analizzato il caso del citizen journalism, per misurare, attraverso la metodologia dell’etnografia digitale, l’entità delle trasformazioni in corso. Il portale di giornalismo partecipativo YouReporter è stato il contesto privilegiato dove poter verificare le ipotesi iniziali circa le dinamiche di partecipazione, e il supporto di programmi di elaborazione statistica netnografica ha permesso di destrutturare al meglio tali dinamiche.

Web advertising. Le comunicazioni commerciali on-line.

La tesi ha ad oggetto lo studio e l’approfondimento delle forme di promozione commerciale presenti in Rete caratterizzate, più che da una normale evoluzione, da continue metamorfosi che ridefiniscono ogni giorno il concetto di pubblicità. L’intento è quello di analizzare il quadro giuridico applicabile alla pubblicità via Web, a fronte della varità di forme e di modalità che essa può assumere. Nel lavoro vengono passate in rassegna le caratteristiche che differenziano la pubblicità commerciale on-line rispetto a quella tradizionale; tra le quali, particolare rilievo assume la capacità d’istaurare una relazione – diretta e non mediata – tra impresa e consumatore. Nel prosieguo viene affrontato il problema dell’individuazione, stante il carattere a-territoriale della Rete, della legge applicabile al web advertising, per poi passare ad una ricognizione delle norme europee ed italiane in materia, senza trascurare quelle emanate in sede di autodisciplina. Ampio spazio è dedicato, infine, all’esame delle diverse e più recenti tecniche di promozione pubblicitaria, di cui sono messi in evidenza gli aspetti tecnico-informatici, imprescindibili ai fini di una corretta valutazione del tema giuridico. In particolare, vengono approfonditi il servizio di posizionamento a pagamento offerto dai principali motori di ricerca (keywords advertising) e gli strumenti di tracciamento dei “comportamenti” on-line degli utenti, che consentono la realizzazione di campagne pubblicitarie mirate (on-line behavioural advertising). Il Web, infatti, non offre più soltanto la possibilità di superare barriere spaziali, linguistiche o temporali e di ampliare la propria sfera di notorietà, ma anche di raggiungere l’utente “interessato” e, pertanto, potenziale acquirente. Di queste nuove realtà pubblicitarie vengono vagliati gli aspetti più critici ed esaminata la disciplina giuridica eventualmente applicabile anche alla luce delle principali decisioni giurisprudenziali nazionali ed europee in materia, nonché delle esperienze giuridiche nord-americane e di tipo autoregolamentare.

Ethical Perspectives on Big Data in Agri-food: Ownership and Governance for Safety

Big data are reshaping the way we interact with technology, thus fostering new applications to increase the safety-assessment of foods. An extraordinary amount of information is analysed using machine learning approaches aimed at detecting the existence or predicting the likelihood of future risks. Food business operators have to share the results of these analyses when applying to place on the market regulated products, whereas agri-food safety agencies (including the European Food Safety Authority) are exploring new avenues to increase the accuracy of their evaluations by processing Big data. Such an informational endowment brings with it opportunities and risks correlated to the extraction of meaningful inferences from data. However, conflicting interests and tensions among the involved entities - the industry, food safety agencies, and consumers - hinder the finding of shared methods to steer the processing of Big data in a sound, transparent and trustworthy way. A recent reform in the EU sectoral legislation, the lack of trust and the presence of a considerable number of stakeholders highlight the need of ethical contributions aimed at steering the development and the deployment of Big data applications. Moreover, Artificial Intelligence guidelines and charters published by European Union institutions and Member States have to be discussed in light of applied contexts, including the one at stake. This thesis aims to contribute to these goals by discussing what principles should be put forward when processing Big data in the context of agri-food safety-risk assessment. The research focuses on two interviewed topics - data ownership and data governance - by evaluating how the regulatory framework addresses the challenges raised by Big data analysis in these domains. The outcome of the project is a tentative Roadmap aimed to identify the principles to be observed when processing Big data in this domain and their possible implementations.

Meaningful insights: explainability techniques for black-box models on tabular data

Artificial Intelligence (AI) and Machine Learning (ML) are novel data analysis techniques providing very accurate prediction results. They are widely adopted in a variety of industries to improve efficiency and decision-making, but they are also being used to develop intelligent systems. Their success grounds upon complex mathematical models, whose decisions and rationale are usually difficult to comprehend for human users to the point of being dubbed as black-boxes. This is particularly relevant in sensitive and highly regulated domains. To mitigate and possibly solve this issue, the Explainable AI (XAI) field became prominent in recent years. XAI consists of models and techniques to enable understanding of the intricated patterns discovered by black-box models. In this thesis, we consider model-agnostic XAI techniques, which can be applied to Tabular data, with a particular focus on the Credit Scoring domain. Special attention is dedicated to the LIME framework, for which we propose several modifications to the vanilla algorithm, in particular: a pair of complementary Stability Indices that accurately measure LIME stability, and the OptiLIME policy which helps the practitioner finding the proper balance among explanations' stability and reliability. We subsequently put forward GLEAMS a model-agnostic surrogate interpretable model which requires to be trained only once, while providing both Local and Global explanations of the black-box model. GLEAMS produces feature attributions and what-if scenarios, from both dataset and model perspective. Eventually, we argue that synthetic data are an emerging trend in AI, being more and more used to train complex models instead of original data. To be able to explain the outcomes of such models, we must guarantee that synthetic data are reliable enough to be able to translate their explanations to real-world individuals. To this end we propose DAISYnt, a suite of tests to measure synthetic tabular data quality and privacy.

Kinematics of local and high-z galaxies through 3D modeling of emission-line datacubes

The kinematics is a fundamental tool to infer the dynamical structure of galaxies and to understand their formation and evolution. Spectroscopic observations of gas emission lines are often used to derive rotation curves and velocity dispersions. It is however difficult to disentangle these two quantities in low spatial-resolution data because of beam smearing. In this thesis, we present 3D-Barolo, a new software to derive the gas kinematics of disk galaxies from emission-line data-cubes. The code builds tilted-ring models in the 3D observational space and compares them with the actual data-cubes. 3D-Barolo works with data at a wide range of spatial resolutions without being affected by instrumental biases. We use 3D-Barolo to derive rotation curves and velocity dispersions of several galaxies in both the local and the high-redshift Universe. We run our code on HI observations of nearby galaxies and we compare our results with 2D traditional approaches. We show that a 3D approach to the derivation of the gas kinematics has to be preferred to a 2D approach whenever a galaxy is resolved with less than about 20 elements across the disk. We moreover analyze a sample of galaxies at z~1, observed in the H-alpha line with the KMOS/VLT spectrograph. Our 3D modeling reveals that the kinematics of these high-z systems is comparable to that of local disk galaxies, with steeply-rising rotation curves followed by a flat part and H-alpha velocity dispersions of 15-40 km/s over the whole disks. This evidence suggests that disk galaxies were already fully settled about 7-8 billion years ago. In summary, 3D-Barolo is a powerful and robust tool to separate physical and instrumental effects and to derive a reliable kinematics. The analysis of large samples of galaxies at different redshifts with 3D-Barolo will provide new insights on how galaxies assemble and evolve throughout cosmic time.