An evolutive approach for smile recognition in video sequences

Facial expression recognition is one of the most challenging research areas in the image recognition ¯eld and has been actively studied since the 70's. For instance, smile recognition has been studied due to the fact that it is considered an important facial expression in human communication, it is therefore likely useful for human–machine interaction. Moreover, if a smile can be detected and also its intensity estimated, it will raise the possibility of new applications in the future

Progetto di reti Sensori Wireless e tecniche di Fusione Sensoriale

Ambient Intelligence (AmI) envisions a world where smart, electronic environments are aware and responsive to their context. People moving into these settings engage many computational devices and systems simultaneously even if they are not aware of their presence. AmI stems from the convergence of three key technologies: ubiquitous computing, ubiquitous communication and natural interfaces. The dependence on a large amount of fixed and mobile sensors embedded into the environment makes of Wireless Sensor Networks one of the most relevant enabling technologies for AmI. WSN are complex systems made up of a number of sensor nodes, simple devices that typically embed a low power computational unit (microcontrollers, FPGAs etc.), a wireless communication unit, one or more sensors and a some form of energy supply (either batteries or energy scavenger modules). Low-cost, low-computational power, low energy consumption and small size are characteristics that must be taken into consideration when designing and dealing with WSNs. In order to handle the large amount of data generated by a WSN several multi sensor data fusion techniques have been developed. The aim of multisensor data fusion is to combine data to achieve better accuracy and inferences than could be achieved by the use of a single sensor alone. In this dissertation we present our results in building several AmI applications suitable for a WSN implementation. The work can be divided into two main areas: Multimodal Surveillance and Activity Recognition. Novel techniques to handle data from a network of low-cost, low-power Pyroelectric InfraRed (PIR) sensors are presented. Such techniques allow the detection of the number of people moving in the environment, their direction of movement and their position. We discuss how a mesh of PIR sensors can be integrated with a video surveillance system to increase its performance in people tracking. Furthermore we embed a PIR sensor within the design of a Wireless Video Sensor Node (WVSN) to extend its lifetime. Activity recognition is a fundamental block in natural interfaces. A challenging objective is to design an activity recognition system that is able to exploit a redundant but unreliable WSN. We present our activity in building a novel activity recognition architecture for such a dynamic system. The architecture has a hierarchical structure where simple nodes performs gesture classification and a high level meta classifiers fuses a changing number of classifier outputs. We demonstrate the benefit of such architecture in terms of increased recognition performance, and fault and noise robustness. Furthermore we show how we can extend network lifetime by performing a performance-power trade-off. Smart objects can enhance user experience within smart environments. We present our work in extending the capabilities of the Smart Micrel Cube (SMCube), a smart object used as tangible interface within a tangible computing framework, through the development of a gesture recognition algorithm suitable for this limited computational power device. Finally the development of activity recognition techniques can greatly benefit from the availability of shared dataset. We report our experience in building a dataset for activity recognition. Such dataset is freely available to the scientific community for research purposes and can be used as a testbench for developing, testing and comparing different activity recognition techniques.

MetaVaccinology: a new vaccine discovery tool

In the last decade, the reverse vaccinology approach shifted the paradigm of vaccine discovery from conventional culture-based methods to high-throughput genome-based approaches for the development of recombinant protein-based vaccines against pathogenic bacteria. Besides reaching its main goal of identifying new vaccine candidates, this new procedure produced also a huge amount of molecular knowledge related to them. In the present work, we explored this knowledge in a species-independent way and we performed a systematic in silico molecular analysis of more than 100 protective antigens, looking at their sequence similarity, domain composition and protein architecture in order to identify possible common molecular features. This meta-analysis revealed that, beside a low sequence similarity, most of the known bacterial protective antigens shared structural/functional Pfam domains as well as specific protein architectures. Based on this, we formulated the hypothesis that the occurrence of these molecular signatures can be predictive of possible protective properties of other proteins in different bacterial species. We tested this hypothesis in Streptococcus agalactiae and identified four new protective antigens. Moreover, in order to provide a second proof of the concept for our approach, we used Staphyloccus aureus as a second pathogen and identified five new protective antigens. This new knowledge-driven selection process, named MetaVaccinology, represents the first in silico vaccine discovery tool based on conserved and predictive molecular and structural features of bacterial protective antigens and not dependent upon the prediction of their sub-cellular localization.

Phylogenomische Analysen bei Metazoen - Zur Stellung der Xenoturbellida und Syndermata

Im Rahmen der vorliegenden Dissertation wurde die phylogenetischen Stellungen der Xenoturbellida (Deuterostomia) und der Syndermata (Protostomia) mit phylogenomischen Techniken untersucht. Auf methodischer Ebene konnte gezeigt werden, dass ribosomale Proteine aufgrund ihres mittleren bis hohen Konservierungsgrades, ihrer Häufigkeit in kleineren EST-Projekten, damit verbunden ihrer Häufigkeit in Datenbanken und ihres phylogenetischen Informationsgehalts nützliche Werkzeuge für phylogenetische Fragestellungen sind. Es konnte durch phylogenetische Rekonstruktionen und Hypothesentests auf Basis eines 11.912 Aminosäuren langen Datensatzes gezeigt werden, dass die Xenoturbellida innerhalb der Deuterostomia eine Schwestergruppenbeziehung zu den Ambulacraria eingehen. Diese Arbeit zeigt im Vergleich aller bisher durchgeführten Arbeiten die beste statistische Unterstützung für diese Topologie. Weiterhin konnte untermauert werden, dass die Urochordata vermutlich anstelle der Cephalochordata die Schwestergruppe der Vertebrata sind. Der Vergleich der publizierten Xenoturbella EST-Datensätze mit dem eigenen Datensatz ließ den Rückschluß zu, dass ESTs offenbar klar weniger anfällig gegen Kontaminationen mit Erbmaterial (DNA+RNA) anderer Spezies sind als PCR-Amplifikate genomischer oder mitochondrialer Gene. Allerdings bestimmt anscheinend der physiologische Zustand der Tiere die Repräsentation von Transkriptklassen wie Stressproteine und mitochondriale Transkripte. Die bakteriellen Transkripte in einem der EST-Datensätze stammen vermutlich von Chlamydien, die möglicherweise symbiontisch in Xenoturbella bocki leben. Im Bereich der Protostomia wurden drei EST-Projekte für Vertreter der Syndermata durchgeführt. Basierend auf drei verschiedenen Proteinalignment-Datensätzen von ca. 11.000 Aminosäuren Länge konnte gezeigt werden, dass die Syndermata innerhalb der Spiralia einzugruppieren sind und dass sie mit den Gnathostomulida das monophyletische Supertaxon Gnathifera bilden. Die genaue phylogenetische Position der Syndermata innerhalb der Spiralia konnte hingegen noch nicht eindeutig geklärt werden, ebenso wie kein kongruenter Beweis für die Existenz des Supertaxons Platyzoa gefunden werden konnte. Im Rahmen der Untersuchung der internen Phylogenie der Syndermata konnten drei der fünf konkurrierenden Hypothesen aufgrund der Paraphylie der Eurotatoria ausgeschlossen werden. Da keine Daten der Seisonidea in den Analysen implementiert waren, bleibt die Frage der internen Phylogenie der Syndermata letztlich offen. Klar ist jedoch, dass die Eurotatoria nicht wie bislang angenommen monophyletisch sind, da die räderorgantragenden Bdelloidea keinesfalls den morphologisch diesbezüglich ähnlichen Monogononta ähnlich sind, sondern den räderorganlosen Acanthocephala näher stehen. Die Abbildung der molekularen Phylogenie auf die morphologischen Verhältnisse zeigt, dass das Räderorgan (partiell oder komplett) offenbar kurz nach der Aufspaltung der Syndermata in Monogononta und Acanthocephala + Bdelloidea in der Acanthocephala + Bdelloidea-Linie reduziert wurde. Die Entstehung des einziehbaren hinteren Körperteils (Rostrum bei Bdelloidea bzw. Proboscis bei Acanthocephala) in der Acanthocephala + Bdelloidea-Linie könnte das Schlüsselereignis zur Entstehung des Endoparasitismus der Acanthocephala gewesen sein.

Between the scales: Water from different perspectives

In der vorliegenden Arbeit werden verschiedene Wassermodelle in sogenannten Multiskalen-Computersimulationen mit zwei Auflösungen untersucht, in atomistischer Auflösung und in einer vergröberten Auflösung, die als "coarse-grained" bezeichnet wird. In der atomistischen Auflösung wird ein Wassermolekül, entsprechend seiner chemischen Struktur, durch drei Atome beschrieben, im Gegensatz dazu wird ein Molekül in der coarse-grained Auflösung durch eine Kugel dargestellt.rnrnDie coarse-grained Modelle, die in dieser Arbeit vorgestellt werden, werden mit verschiedenen coarse-graining Methoden entwickelt. Hierbei kommen hauptsächlich die "iterative Boltzmann Inversion" und die "iterative Monte Carlo Inversion" zum Einsatz. Beides sind struktur-basierte Ansätze, die darauf abzielen bestimmte strukturelle Eigenschaften, wie etwa die Paarverteilungsfunktionen, des zugrundeliegenden atomistischen Systems zu reproduzieren. Zur automatisierten Anwendung dieser Methoden wurde das Softwarepaket "Versatile Object-oriented Toolkit for Coarse-Graining Applications" (VOTCA) entwickelt.rnrnEs wird untersucht, in welchem Maße coarse-grained Modelle mehrere Eigenschaftenrndes zugrundeliegenden atomistischen Modells gleichzeitig reproduzieren können, z.B. thermodynamische Eigenschaften wie Druck und Kompressibilität oder strukturelle Eigenschaften, die nicht zur Modellbildung verwendet wurden, z.B. das tetraedrische Packungsverhalten, welches für viele spezielle Eigenschaft von Wasser verantwortlich ist.rnrnMit Hilfe des "Adaptive Resolution Schemes" werden beide Auflösungen in einer Simulation kombiniert. Dabei profitiert man von den Vorteilen beider Modelle:rnVon der detaillierten Darstellung eines räumlich kleinen Bereichs in atomistischer Auflösung und von der rechnerischen Effizienz des coarse-grained Modells, die den Bereich simulierbarer Zeit- und Längenskalen vergrössert.rnrnIn diesen Simulationen kann der Einfluss des Wasserstoffbrückenbindungsnetzwerks auf die Hydration von Fullerenen untersucht werden. Es zeigt sich, dass die Struktur der Wassermoleküle an der Oberfläche hauptsächlich von der Art der Wechselwirkung zwischen dem Fulleren und Wasser und weniger von dem Wasserstoffbrückenbindungsnetzwerk dominiert wird.rn

Applicazione mobile per supporto alla didattica

Questa tesi descrive le fasi della progettazione e dell'implementazione di una applicazione mobile per il supporto alla didattica del corso di Programmazione del CdL di Ingegneria e Scienze Informatiche del Campus di Cesena. Il progetto ha lo scopo di mettere a disposizione degli studenti e dei docenti una applicazione per dispositivi Android che permetta di usufruire dei servizi attualmente forniti dal portale ufficiale del corso. Le funzionalità principali consistono nell'accesso ai materiali didattici e nella gestione delle consegne di esercizi propedeutici alla prova finale per quanto concerne gli studenti; ai professori è invece fornita la possibilità di eseguire la correzione degli elaborati e seguirne lo stato di avanzamento. Visto l'elevato numero di utenti che utilizzava il portale esistente tramite tablet e smartphone si è ritenuto necessario lo sviluppo di questo prodotto che possa fornire una user experience ottimizzata per questi dispositivi. Durante la progettazione è stata data particolare importanza all'ottimizzazione delle prestazioni, sfruttando gli strumenti più recenti forniti agli sviluppatori, e alla conformità con i principali design pattern della programmazione per dispositivi Android. Allo scopo di fornire un prodotto utilizzabile dalla maggior parte dell'utenza potenziale, si è inoltre data priorità alla compatibilità con tutti i dispositivi e le versioni del sistema operativo, senza rinunciare ad alcuna funzionalità. Il risultato del lavoro consiste in un prototipo pienamente funzionante e utilizzabile che mira a fornire una base stabile su cui eseguire future evoluzioni.

Decomposition Methods and Network Design Problems

Decomposition based approaches are recalled from primal and dual point of view. The possibility of building partially disaggregated reduced master problems is investigated. This extends the idea of aggregated-versus-disaggregated formulation to a gradual choice of alternative level of aggregation. Partial aggregation is applied to the linear multicommodity minimum cost flow problem. The possibility of having only partially aggregated bundles opens a wide range of alternatives with different trade-offs between the number of iterations and the required computation for solving it. This trade-off is explored for several sets of instances and the results are compared with the ones obtained by directly solving the natural node-arc formulation. An iterative solution process to the route assignment problem is proposed, based on the well-known Frank Wolfe algorithm. In order to provide a first feasible solution to the Frank Wolfe algorithm, a linear multicommodity min-cost flow problem is solved to optimality by using the decomposition techniques mentioned above. Solutions of this problem are useful for network orientation and design, especially in relation with public transportation systems as the Personal Rapid Transit. A single-commodity robust network design problem is addressed. In this, an undirected graph with edge costs is given together with a discrete set of balance matrices, representing different supply/demand scenarios. The goal is to determine the minimum cost installation of capacities on the edges such that the flow exchange is feasible for every scenario. A set of new instances that are computationally hard for the natural flow formulation are solved by means of a new heuristic algorithm. Finally, an efficient decomposition-based heuristic approach for a large scale stochastic unit commitment problem is presented. The addressed real-world stochastic problem employs at its core a deterministic unit commitment planning model developed by the California Independent System Operator (ISO).

Coating of surfaces of composite materials with enzymatically formed biosilica

Die Bildung kieselsäurehaltiger Spicula in marinen Schwämmen ist nur möglich durch die enzymatische Aktivität des Silicatein- in Verbindung mit der stöchiometrischen Selbstassemblierung des Enzyms mit anderen Schwammproteinen. Die vorliegende Arbeit basiert auf einem biomimetischen Ansatz mit dem Ziel, unterschiedliche Oberflächen für biotechnologische und biomedizinische Anwendungen mit Biosilica und Biotitania zu beschichten und zu funktionalisieren. Für biotechnologische Anwendungen ist dabei das Drucken von Cystein-getaggtem Silicatein auf Gold-Oberflächen von Bedeutung, denn es ermöglichte die Bildung definierter Biotitania-Strukturen (Anatas), welche als Photokatalysator den Abbau eines organischen Farbstoffs bewirkten. Des Weiteren zeigte sich die bio-inspirierte Modifikation von Tyrosin-Resten an rekombinantem Silicatein-(via Tyrosinase) als vielversprechendes Werkzeug zur Beschleunigung der Selbstassemblierung des Enzyms zu mesoskaligen Filamenten. Durch eine solche Modifikation konnte Silicatein auch auf der Oberfläche von anorganischen Partikeln immobilisiert werden, welches die Assemblierung von anorganisch-organischen Verbundwerkstoffen in wäßriger Umgebung förderte. Die resultierenden supramolekularen Strukturen könnten dabei in bio-inspirierten und biotechnologischen Anwendungen genutzt werden. Weiterhin wurde in der vorliegenden Arbeit die Sekundärstruktur von rekombinantem Silicatein- (Monomer und Oligomer) durch Raman Spektroskopie analysiert, nachdem das Protein gemäß einer neu etablierten Methode rückgefaltet worden war. Diese Spektraldaten zeigten insbesondere Änderungen der Proteinkonformation durch Solubilisierung und Oligomerisierung des Enzyms. Außerdem wurden die osteoinduzierenden und osteogenen Eigenschaften unterschiedlicher organischer Polymere, die herkömmlich als Knochenersatzmaterial genutzt werden, durch Oberflächenmodifikation mit Silicatein/Biosilica verbessert: Die bei der Kultivierung knochenbildender Zellen auf derart oberflächenbehandelten Materialien beobachtete verstärkte Biomineralisierung, Aktivierung der Alkalischen Phosphatase, und Ausbildung eines typischen zellulären Phänotyps verdeutlichen das Potential von Silicatein/Biosilica für der Herstellung neuartiger Implantat- und Knochenersatzmaterialien.

Progettazione e realizzazione di soluzioni per la geolocalizzazione e marketing di prossimità mediante utilizzo di tecnologia iBeacons.

Si analizza il nuovo protocollo iBeacon, proposto da Apple, e come questo possa essere utilizzato per migliorare e rendere maggiormente efficace la user experience delle applicazioni. Viene inoltre definito lo stato dell'arte di questa tecnologia analizzando l'hardware e il software a oggi presenti sul mercato. Proponendo un esempio di come tale protocollo possa essere utilizzato per implementare tecniche di marketing di prossimità.

Interfaccia utente in applicazione mobile con analisi dell'usability, dei workflow e del contesto operativo dell'utente

In questo lavoro di tesi verrà presentato un applicativo, sviluppato con l’azienda EBWorld, per dispositivi con sistema operativo Android. L’applicazione ha come destinatari i tecnici e gli operatori sul campo di aziende clienti di EBWorld. Nel dispositivo vengono caricati i dati estratti dal database (porzioni di mappe e informazioni ad esse correlate) che vengono lette e mostrate nello schermo. Le funzionalità fornite sono: utilizzo dello strumento trail, per effettuare misurazioni; creazione di progetti all’interno delle esportazioni; inserimento di sketch, definiti in accordo con l’azienda, all’interno dei progetti; selezione degli sketch e delle informazioni estratte dal database e visualizzazione delle relative informazioni / proprietà; eliminazione di sketch inseriti. È stato effettuato uno studio di progettazione dell’interfaccia per offrire un’ottima usabilità anche in situazioni critiche.

Sviluppo di applicazioni mobili attraverso framework cross-platform: una valutazione sperimentale

Sviluppare e manutenere applicativi destinati a differenti piattaforme è un’opzione esclusiva di quelle entità capaci di sostenere costi molto alti per la realizzazione di queste applicazioni. Questo esclude gli sviluppatori indipendenti, che spesso realizzano prodotti in totale autonomia; le start-up, che hanno l’esigenza di sviluppare un’idea di business avendo a disposizione budget estremamente ridotti; le piccole aziende, alle quali viene così preclusa la possibilità di competere con player più importanti. Questo tipo di emergenze rende lo sviluppo cross-platform una soluzione interessante per la realizzazione delle applicazioni mobili, abbattendo i costi di sviluppo e permettendo di raggiungere più velocemente un pubblico più ampio. C'è quindi sempre maggiore interesse, da parte degli sviluppatori, per gli strumenti di sviluppo cross-platform. Per catturare l’attenzione degli sviluppatori è necessario che lo strumento sia dotato di buona stabilità, che offra un ambiente di sviluppo confortevole, una buona user experience, facilità di aggiornamento, tempi di sviluppo contenuti e possibilità di immissione delle applicazioni su diversi ecosistemi software. L’idea alla base di questa Tesi di laurea è valutare i pro e i contro di uno di questi framework cross-platform e compararlo con le tecnologie native. Il framework scelto è Ionic per via della sua popolarità tra gli sviluppatori e della ridotta bibliografia scientifica a riguardo. Molte ricerche scientifiche valutano le prestazioni di uno o più framework cross-platform rispetto ad una soluzione nativa, tuttavia è raro un confronto tra un framework e più soluzioni native. Per questo, oltre a valutare i pro e i contro del framework, verrà anche effettuata una comparazione tra gli applicativi nativi per Android e iOS e le controparti sviluppate attraverso Ionic, permettendo di identificare eventuali differenze di performance e aiutare gli sviluppatori nelle scelte tecnologiche.

Spatio-temporal credit assignment in population learning

Learning by reinforcement is important in shaping animal behavior, and in particular in behavioral decision making. Such decision making is likely to involve the integration of many synaptic events in space and time. However, using a single reinforcement signal to modulate synaptic plasticity, as suggested in classical reinforcement learning algorithms, a twofold problem arises. Different synapses will have contributed differently to the behavioral decision, and even for one and the same synapse, releases at different times may have had different effects. Here we present a plasticity rule which solves this spatio-temporal credit assignment problem in a population of spiking neurons. The learning rule is spike-time dependent and maximizes the expected reward by following its stochastic gradient. Synaptic plasticity is modulated not only by the reward, but also by a population feedback signal. While this additional signal solves the spatial component of the problem, the temporal one is solved by means of synaptic eligibility traces. In contrast to temporal difference (TD) based approaches to reinforcement learning, our rule is explicit with regard to the assumed biophysical mechanisms. Neurotransmitter concentrations determine plasticity and learning occurs fully online. Further, it works even if the task to be learned is non-Markovian, i.e. when reinforcement is not determined by the current state of the system but may also depend on past events. The performance of the model is assessed by studying three non-Markovian tasks. In the first task, the reward is delayed beyond the last action with non-related stimuli and actions appearing in between. The second task involves an action sequence which is itself extended in time and reward is only delivered at the last action, as it is the case in any type of board-game. The third task is the inspection game that has been studied in neuroeconomics, where an inspector tries to prevent a worker from shirking. Applying our algorithm to this game yields a learning behavior which is consistent with behavioral data from humans and monkeys, revealing themselves properties of a mixed Nash equilibrium. The examples show that our neuronal implementation of reward based learning copes with delayed and stochastic reward delivery, and also with the learning of mixed strategies in two-opponent games.

Spatio-temporal credit assignment in population learning

Learning by reinforcement is important in shaping animal behavior. But behavioral decision making is likely to involve the integration of many synaptic events in space and time. So in using a single reinforcement signal to modulate synaptic plasticity a twofold problem arises. Different synapses will have contributed differently to the behavioral decision and, even for one and the same synapse, releases at different times may have had different effects. Here we present a plasticity rule which solves this spatio-temporal credit assignment problem in a population of spiking neurons. The learning rule is spike time dependent and maximizes the expected reward by following its stochastic gradient. Synaptic plasticity is modulated not only by the reward but by a population feedback signal as well. While this additional signal solves the spatial component of the problem, the temporal one is solved by means of synaptic eligibility traces. In contrast to temporal difference based approaches to reinforcement learning, our rule is explicit with regard to the assumed biophysical mechanisms. Neurotransmitter concentrations determine plasticity and learning occurs fully online. Further, it works even if the task to be learned is non-Markovian, i.e. when reinforcement is not determined by the current state of the system but may also depend on past events. The performance of the model is assessed by studying three non-Markovian tasks. In the first task the reward is delayed beyond the last action with non-related stimuli and actions appearing in between. The second one involves an action sequence which is itself extended in time and reward is only delivered at the last action, as is the case in any type of board-game. The third is the inspection game that has been studied in neuroeconomics. It only has a mixed Nash equilibrium and exemplifies that the model also copes with stochastic reward delivery and the learning of mixed strategies.

Spatio-temporal credit assignment in population learning

We present a model for plasticity induction in reinforcement learning which is based on a cascade of synaptic memory traces. In the cascade of these so called eligibility traces presynaptic input is first corre lated with postsynaptic events, next with the behavioral decisions and finally with the external reinforcement. A population of leaky integrate and fire neurons endowed with this plasticity scheme is studied by simulation on different tasks. For operant co nditioning with delayed reinforcement, learning succeeds even when the delay is so large that the delivered reward reflects the appropriateness, not of the immediately preceeding response, but of a decision made earlier on in the stimulus - decision sequence . So the proposed model does not rely on the temporal contiguity between decision and pertinent reward and thus provides a viable means of addressing the temporal credit assignment problem. In the same task, learning speeds up with increasing population si ze, showing that the plasticity cascade simultaneously addresses the spatial problem of assigning credit to the different population neurons. Simulations on other task such as sequential decision making serve to highlight the robustness of the proposed sch eme and, further, contrast its performance to that of temporal difference based approaches to reinforcement learning.

Percutaneous and Open Retrograde Endovascular Stenting of Symptomatic High-Grade Innominate Artery Stenosis: Technique and Follow-Up

Angioplasty and stenting of the IA have been reported with high technical and clinical success rates, low complication rates and good mid-term patency rates. Different antegrade or retrograde endovascular catheter-based approaches and combinations with surgical exposure of the CCA are used. The purpose of this study was to determine safety, efficacy and mid-term clinical and radiological outcome of the stent-assisted treatment of atherosclerotic stenotic disease of the IA with special focus on the different technical approaches.