Sistemi hands-free basati su smart-glass a supporto di attivita di customer care di un'impresa

Da alcuni anni in ambito business ed enterprise si sta diffondendo l'utilizzo di dispositivi wearable al fine di rendere più efficiente ed efficace la gestione di molteplici attività e processi aziendali. I sistemi hand-held comunemente utilizzati in ambito lavorativo, tra cui smartphone e tablet, spesso non risultano idonei in contesti in cui un operatore debba interagire con il dispositivo mentre ha le proprie mani impegnate con attrezzature e strumenti di lavoro. I sistemi hands-free rimediano a tali problematiche supportando tecniche di interazione non convenzionali che consentono all'operatore di mantenere libere le proprie mani durante la consultazione del dispositivo e di rimanere concentrato sull'attività che sta svolgendo. I sistemi basati su smart-glass, oltre ad offrire funzionalità hands-free, presentano l'ulteriore vantaggio di poter presentare all'interno del campo visivo dell'utente importanti informazioni di supporto inerenti all'attività che sta svolgendo, avvalendosi anche dell'utilizzo di tecnologie di realtà aumentata. La sinergia tra dispositivi basati su smart-glass e tecniche di realtà aumentata sta destando un crescente interesse sia in ambito accademico che industriale; esiste la possibilità che in un prossimo futuro questa tipologia di sistemi divenga la nuova piattaforma computazionale enterprise di riferimento. L'obiettivo di questo lavoro di tesi è stato lo studio e la progettazione di una soluzione hands-free basata su smart-glass in grado di supportare alcune attività di customer care del Gruppo Loccioni, una società che si occupa dello sviluppo di sistemi automatici di misura e controllo per migliorare la qualità, l'efficienza e la sostenibilità di prodotti, processi ed edifici. In particolare, il sistema sviluppato ha consentito di migliorare la gestione dei processi di manutenzione e riparazione degli impianti energetici sostenibili che il Gruppo Loccioni installa presso le sedi di imprese clienti.

Explicit context matching in content-based publish/subscribe systems

Although context could be exploited to improve performance, elasticity and adaptation in most distributed systems that adopt the publish/subscribe (P/S) communication model, only a few researchers have focused on the area of context-aware matching in P/S systems and have explored its implications in domains with highly dynamic context like wireless sensor networks (WSNs) and IoT-enabled applications. Most adopted P/S models are context agnostic or do not differentiate context from the other application data. In this article, we present a novel context-aware P/S model. SilboPS manages context explicitly, focusing on the minimization of network overhead in domains with recurrent context changes related, for example, to mobile ad hoc networks (MANETs). Our approach represents a solution that helps to efficiently share and use sensor data coming from ubiquitous WSNs across a plethora of applications intent on using these data to build context awareness. Specifically, we empirically demonstrate that decoupling a subscription from the changing context in which it is produced and leveraging contextual scoping in the filtering process notably reduces (un)subscription cost per node, while improving the global performance/throughput of the network of brokers without fltering the cost of SIENA-like topology changes.

The role of context in image annotation and recommendation

With the rise of smart phones, lifelogging devices (e.g. Google Glass) and popularity of image sharing websites (e.g. Flickr), users are capturing and sharing every aspect of their life online producing a wealth of visual content. Of these uploaded images, the majority are poorly annotated or exist in complete semantic isolation making the process of building retrieval systems difficult as one must firstly understand the meaning of an image in order to retrieve it. To alleviate this problem, many image sharing websites offer manual annotation tools which allow the user to “tag” their photos, however, these techniques are laborious and as a result have been poorly adopted; Sigurbjörnsson and van Zwol (2008) showed that 64% of images uploaded to Flickr are annotated with < 4 tags. Due to this, an entire body of research has focused on the automatic annotation of images (Hanbury, 2008; Smeulders et al., 2000; Zhang et al., 2012a) where one attempts to bridge the semantic gap between an image’s appearance and meaning e.g. the objects present. Despite two decades of research the semantic gap still largely exists and as a result automatic annotation models often offer unsatisfactory performance for industrial implementation. Further, these techniques can only annotate what they see, thus ignoring the “bigger picture” surrounding an image (e.g. its location, the event, the people present etc). Much work has therefore focused on building photo tag recommendation (PTR) methods which aid the user in the annotation process by suggesting tags related to those already present. These works have mainly focused on computing relationships between tags based on historical images e.g. that NY and timessquare co-exist in many images and are therefore highly correlated. However, tags are inherently noisy, sparse and ill-defined often resulting in poor PTR accuracy e.g. does NY refer to New York or New Year? This thesis proposes the exploitation of an image’s context which, unlike textual evidences, is always present, in order to alleviate this ambiguity in the tag recommendation process. Specifically we exploit the “what, who, where, when and how” of the image capture process in order to complement textual evidences in various photo tag recommendation and retrieval scenarios. In part II, we combine text, content-based (e.g. # of faces present) and contextual (e.g. day-of-the-week taken) signals for tag recommendation purposes, achieving up to a 75% improvement to precision@5 in comparison to a text-only TF-IDF baseline. We then consider external knowledge sources (i.e. Wikipedia & Twitter) as an alternative to (slower moving) Flickr in order to build recommendation models on, showing that similar accuracy could be achieved on these faster moving, yet entirely textual, datasets. In part II, we also highlight the merits of diversifying tag recommendation lists before discussing at length various problems with existing automatic image annotation and photo tag recommendation evaluation collections. In part III, we propose three new image retrieval scenarios, namely “visual event summarisation”, “image popularity prediction” and “lifelog summarisation”. In the first scenario, we attempt to produce a rank of relevant and diverse images for various news events by (i) removing irrelevant images such memes and visual duplicates (ii) before semantically clustering images based on the tweets in which they were originally posted. Using this approach, we were able to achieve over 50% precision for images in the top 5 ranks. In the second retrieval scenario, we show that by combining contextual and content-based features from images, we are able to predict if it will become “popular” (or not) with 74% accuracy, using an SVM classifier. Finally, in chapter 9 we employ blur detection and perceptual-hash clustering in order to remove noisy images from lifelogs, before combining visual and geo-temporal signals in order to capture a user’s “key moments” within their day. We believe that the results of this thesis show an important step towards building effective image retrieval models when there lacks sufficient textual content (i.e. a cold start).

Recommendation and planning through mobile devices in tourism context

In this paper we present a mobile recommendation and planning system, named PSiS Mobile. It is designed to provide effective support during a tourist visit through context-aware information and recommendations about points of interest, exploiting tourist preferences and context. Designing a tool like this brings several challenges that must be addressed. We discuss how these challenges have been overcame, present the overall system architecture, since this mobile application extends the PSiS project website, and the mobile application architecture.

Context classification for service robots

This dissertation presents a solution for environment sensing using sensor fusion techniques and a context/environment classification of the surroundings in a service robot, so it could change his behavior according to the different rea-soning outputs. As an example, if a robot knows he is outdoors, in a field environment, there can be a sandy ground, in which it should slow down. Contrariwise in indoor environments, that situation is statistically unlikely to happen (sandy ground). This simple assumption denotes the importance of context-aware in automated guided vehicles.

O âmbito de aplicação material da legislação de proteção de dados pessoais e o tratamento de imagem por pessoa singular

Images have gained a never before seen importance. Technological changes have given the Information Society extraordinary means to capture, treat and transmit images, wheter your own or those of others, with or without a commercial purpose, with no boundaries of time or country, without “any kind of eraser”. From the several different ways natural persons may engage in image processing with no commercial purpose, the cases of sharing pictures through social networks and video surveillance assume particular relevance. Consequently there are growing legitimate concerns with the protection of one's image, since its processing may sometimes generate situations of privacy invasion or put at risk other fundamental rights. With this in mind, the present thesis arises from the question: what are the existent legal instruments in Portuguese Law that enable citizens to protect themselves from the abusive usage of their own pictures, whether because that image have been captured by a smartphone or some video surveillance camera, whether because it was massively shared through a blog or some social network? There is no question the one's right to not having his or her image used in an abusive way is protected by the Portuguese constitution, through the article 26th CRP, as well as personally right, under the article 79th of the Civil Code, and finally through criminal law, articles 192nd and 193rd of the Criminal Code. The question arises in the personal data protection context, considering that one's picture, given certain conditions, is personal data. Both the Directive 95/46/CE dated from 1995 as well as the LPD from 1998 are applicable to the processing of personal data, but both exclude situations of natural persons doing so in the pursuit of activities strictly personal or family-related. These laws demand complex procedures to natural persons, such as the preemptive formal authorisation request to the Data Protection National Commission. Failing to do so a natural person may result in the application of fines as high as €2.500,00 or even criminal charges. Consequently, the present thesis aims to study if the image processing with no commercial purposes by a natural person in the context of social networks or through video surveillance belongs to the domain of the existent personal data protection law. To that effect, it was made general considerations regarding the concept of video surveillance, what is its regimen, in a way that it may be distinguishable from Steve Mann's definition of sousveillance, and what are the associated obligations in order to better understand the concept's essence. The application of the existent laws on personal data protection to images processing by natural persons has been analysed taking into account the Directive 95/46/CE, the LPD and the General Regulation. From this analysis it is concluded that the regimen from 1995 to 1998 is out of touch with reality creating an absence of legal shielding in the personal data protection law, a flaw that doesn't exist because compensated by the right to image as a right to personality, that anyway reveals the inability of the Portuguese legislator to face the new technological challenges. It is urgent to legislate. A contrary interpretation will evidence the unconstitutionality of several rules on the LPD due to the obligations natural persons are bound to that violate the right to the freedom of speech and information, which would be inadequate and disproportionate. Considering the recently approved General Regulation and in the case it becomes the final version, the use for natural person of video surveillance of private spaces, Google Glass (in public and private places) and other similar gadgets used to recreational purposes, as well as social networks are subject to its regulation only if the images are shared without limits or existing commercial purposes. Video surveillance of public spaces in all situations is subject to General Regulation provisions.

Design and Implementation of MUPE Server Context Tool for Eclipse IDE

The development of software tools begun as the first computers were built. The current generation of development environments offers a common interface to access multiple software tools and often also provide a possibility to build custom tools as extensions to the existing development environment. Eclipse is an open source development environment that offers good starting point for developing custom extensions. This thesis presents a software tool to aid the development of context-aware applications on Multi-User Publishing Environment (MUPE) platform. The tool is implemented as an Eclipse plug-in. The tool allows developer to include external server side contexts to their MUPE applications. The tool allows additional context sources to be added through the Eclipse's extension point mechanism. The thesis describes how the tool was designed and implemented. The implementation consists of tool core component part and an additional context source extension part. Tool core component is responsible for the actual context addition and also provides the needed user interface elements to the Eclipse workbench. Context source component provides the needed context source related information to the core component. As part of the work an update site feature was also implemented for distributing the tool through Eclipse update mechanism.

Trustworthy context dependency in ubiquitous systems

The modern society is getting increasingly dependent on software applications. These run on processors, use memory and account for controlling functionalities that are often taken for granted. Typically, applications adjust the functionality in response to a certain context that is provided or derived from the informal environment with various qualities. To rigorously model the dependence of an application on a context, the details of the context are abstracted and the environment is assumed stable and fixed. However, in a context-aware ubiquitous computing environment populated by autonomous agents, a context and its quality parameters may change at any time. This raises the need to derive the current context and its qualities at runtime. It also implies that a context is never certain and may be subjective, issues captured by the context’s quality parameter of experience-based trustworthiness. Given this, the research question of this thesis is: In what logical topology and by what means may context provided by autonomous agents be derived and formally modelled to serve the context-awareness requirements of an application? This research question also stipulates that the context derivation needs to incorporate the quality of the context. In this thesis, we focus on the quality of context parameter of trustworthiness based on experiences having a level of certainty and referral experiences, thus making trustworthiness reputation based. Hence, in this thesis we seek a basis on which to reason and analyse the inherently inaccurate context derived by autonomous agents populating a ubiquitous computing environment in order to formally model context-awareness. More specifically, the contribution of this thesis is threefold: (i) we propose a logical topology of context derivation and a method of calculating its trustworthiness, (ii) we provide a general model for storing experiences and (iii) we formalise the dependence between the logical topology of context derivation and its experience-based trustworthiness. These contributions enable abstraction of a context and its quality parameters to a Boolean decision at runtime that may be formally reasoned with. We employ the Action Systems framework for modelling this. The thesis is a compendium of the author’s scientific papers, which are republished in Part II. Part I introduces the field of research by providing the mending elements for the thesis to be a coherent introduction for addressing the research question. In Part I we also review a significant body of related literature in order to better illustrate our contributions to the research field.

Context as a Service

In the vision of Mark Weiser on ubiquitous computing, computers are disappearing from the focus of the users and are seamlessly interacting with other computers and users in order to provide information and services. This shift of computers away from direct computer interaction requires another way of applications to interact without bothering the user. Context is the information which can be used to characterize the situation of persons, locations, or other objects relevant for the applications. Context-aware applications are capable of monitoring and exploiting knowledge about external operating conditions. These applications can adapt their behaviour based on the retrieved information and thus to replace (at least a certain amount) the missing user interactions. Context awareness can be assumed to be an important ingredient for applications in ubiquitous computing environments. However, context management in ubiquitous computing environments must reflect the specific characteristics of these environments, for example distribution, mobility, resource-constrained devices, and heterogeneity of context sources. Modern mobile devices are equipped with fast processors, sufficient memory, and with several sensors, like Global Positioning System (GPS) sensor, light sensor, or accelerometer. Since many applications in ubiquitous computing environments can exploit context information for enhancing their service to the user, these devices are highly useful for context-aware applications in ubiquitous computing environments. Additionally, context reasoners and external context providers can be incorporated. It is possible that several context sensors, reasoners and context providers offer the same type of information. However, the information providers can differ in quality levels (e.g. accuracy), representations (e.g. position represented in coordinates and as an address) of the offered information, and costs (like battery consumption) for providing the information. In order to simplify the development of context-aware applications, the developers should be able to transparently access context information without bothering with underlying context accessing techniques and distribution aspects. They should rather be able to express which kind of information they require, which quality criteria this information should fulfil, and how much the provision of this information should cost (not only monetary cost but also energy or performance usage). For this purpose, application developers as well as developers of context providers need a common language and vocabulary to specify which information they require respectively they provide. These descriptions respectively criteria have to be matched. For a matching of these descriptions, it is likely that a transformation of the provided information is needed to fulfil the criteria of the context-aware application. As it is possible that more than one provider fulfils the criteria, a selection process is required. In this process the system has to trade off the provided quality of context and required costs of the context provider against the quality of context requested by the context consumer. This selection allows to turn on context sources only if required. Explicitly selecting context services and thereby dynamically activating and deactivating the local context provider has the advantage that also the resource consumption is reduced as especially unused context sensors are deactivated. One promising solution is a middleware providing appropriate support in consideration of the principles of service-oriented computing like loose coupling, abstraction, reusability, or discoverability of context providers. This allows us to abstract context sensors, context reasoners and also external context providers as context services. In this thesis we present our solution consisting of a context model and ontology, a context offer and query language, a comprehensive matching and mediation process and a selection service. Especially the matching and mediation process and the selection service differ from the existing works. The matching and mediation process allows an autonomous establishment of mediation processes in order to transfer information from an offered representation into a requested representation. In difference to other approaches, the selection service selects not only a service for a service request, it rather selects a set of services in order to fulfil all requests which also facilitates the sharing of services. The approach is extensively reviewed regarding the different requirements and a set of demonstrators shows its usability in real-world scenarios.

Context detection and abstraction in smart environments

Context-aware computing is currently considered the most promising approach to overcome information overload and to speed up access to relevant information and services. Context-awareness may be derived from many sources, including user profile and preferences, network information, sensor analysis; usually context-awareness relies on the ability of computing devices to interact with the physical world, i.e. with the natural and artificial objects hosted within the "environment”. Ideally, context-aware applications should not be intrusive and should be able to react according to user’s context, with minimum user effort. Context is an application dependent multidimensional space and the location is an important part of it since the very beginning. Location can be used to guide applications, in providing information or functions that are most appropriate for a specific position. Hence location systems play a crucial role. There are several technologies and systems for computing location to a vary degree of accuracy and tailored for specific space model, i.e. indoors or outdoors, structured spaces or unstructured spaces. The research challenge faced by this thesis is related to pedestrian positioning in heterogeneous environments. Particularly, the focus will be on pedestrian identification, localization, orientation and activity recognition. This research was mainly carried out within the “mobile and ambient systems” workgroup of EPOCH, a 6FP NoE on the application of ICT to Cultural Heritage. Therefore applications in Cultural Heritage sites were the main target of the context-aware services discussed. Cultural Heritage sites are considered significant test-beds in Context-aware computing for many reasons. For example building a smart environment in museums or in protected sites is a challenging task, because localization and tracking are usually based on technologies that are difficult to hide or harmonize within the environment. Therefore it is expected that the experience made with this research may be useful also in domains other than Cultural Heritage. This work presents three different approaches to the pedestrian identification, positioning and tracking: Pedestrian navigation by means of a wearable inertial sensing platform assisted by the vision based tracking system for initial settings an real-time calibration; Pedestrian navigation by means of a wearable inertial sensing platform augmented with GPS measurements; Pedestrian identification and tracking, combining the vision based tracking system with WiFi localization. The proposed localization systems have been mainly used to enhance Cultural Heritage applications in providing information and services depending on the user’s actual context, in particular depending on the user’s location.

Middleware for quality-based context distribution in mobile systems

The continuous advancements and enhancements of wireless systems are enabling new compelling scenarios where mobile services can adapt according to the current execution context, represented by the computational resources available at the local device, current physical location, people in physical proximity, and so forth. Such services called context-aware require the timely delivery of all relevant information describing the current context, and that introduces several unsolved complexities, spanning from low-level context data transmission up to context data storage and replication into the mobile system. In addition, to ensure correct and scalable context provisioning, it is crucial to integrate and interoperate with different wireless technologies (WiFi, Bluetooth, etc.) and modes (infrastructure-based and ad-hoc), and to use decentralized solutions to store and replicate context data on mobile devices. These challenges call for novel middleware solutions, here called Context Data Distribution Infrastructures (CDDIs), capable of delivering relevant context data to mobile devices, while hiding all the issues introduced by data distribution in heterogeneous and large-scale mobile settings. This dissertation thoroughly analyzes CDDIs for mobile systems, with the main goal of achieving a holistic approach to the design of such type of middleware solutions. We discuss the main functions needed by context data distribution in large mobile systems, and we claim the precise definition and clean respect of quality-based contracts between context consumers and CDDI to reconfigure main middleware components at runtime. We present the design and the implementation of our proposals, both in simulation-based and in real-world scenarios, along with an extensive evaluation that confirms the technical soundness of proposed CDDI solutions. Finally, we consider three highly heterogeneous scenarios, namely disaster areas, smart campuses, and smart cities, to better remark the wide technical validity of our analysis and solutions under different network deployments and quality constraints.

Realtà aumentata per dispositivi android: lo stato dell'arte

Obiettivo di questa tesi è quello di illustrare il mondo della realtà aumentata (AR) ed in particolare delle tecnologie software disponibili per lo sviluppo di applicazioni su dispositivi Android. Si partirà dal darne una definizione e riassumerne i principali fatti storici, all'illustrarne i vari hardware disponibili sul mercato e le tecnologie software per sviluppare progetti. Non verranno tralasciati utilizzi e settori di ricerca, e si presenterà poi il sistema operativo Android. Dopo uno sguardo alla sua architettura e alle sue caratteristiche, nonché al linguaggio di programmazione Java, cardine per lo sviluppo in questo sistema, si presenteranno alcune API dell'SDK nativo che si rivelano utili per lo sviluppo di applicazioni per la realtà aumentata. Infine, verrà presentato un approfondimento sull'SDK Metaio.

Un Sistema di Parcheggio Intelligente basato su Context-Awareness

Sistema di segnalazione automatica posti auto su strada. Implementato per Android con tecniche di apprendimento automatico supervisionato e Bluetooth per realizzare un'applicazione Context-Aware.

Realtà Aumentata applicata alla manutenzione tecnica di macchinari industriali

La tesi, svolta per il completamento della Laurea Magistrale in Ingegneria Informatica, tratta la realizzazione di un progetto prototipo di Computer Vision (CV) e Realtà Aumentata (RA) per la manutenzione tecnica di macchinari industriali attraverso l'utilizzo di dispositivi mobili See-Through. Lo scopo è stato, oltre lo studio dello stato dell'arte in materia, provare con mano e rendere maggiormente visibili al pubblico questi nuovi rami dell'informatica. Il prototipo creato è stato inserito in un contesto aziendale, con misurazioni e prove sul campo. Partendo da una breve introduzione sulla realtà aumentata, nel primo capitolo viene descritto il progetto sviluppato, diviso in due sottoprogetti. Il primo, svolto solamente in una fase iniziale e presentato nel secondo capitolo, espone la realizzazione di un'applicazione mobile per lo streaming video con l'aggiunta di contenuti grafici aumentati. Il secondo, progettato e sviluppato in totale autonomia, rappresenta un prototipo demo di utilizzo della RA. La realizzazione viene illustrata nei capitoli successivi. Nel terzo capitolo si introducono gli strumenti che sono stati utilizzati per lo sviluppo dell'applicazione, in particolare Unity (per il development multi-piattaforma), Vuforia (per gli algoritmi di CV) e Blender (per la realizzazione di procedure di manutenzione). Il quarto capitolo, la parte più rilevante della trattazione, descrive, passo dopo passo, la creazione dei vari componenti, riassumendo in modo conciso e attraverso l'uso di figure i punti cardine. Infine, il quinto capitolo conclude il percorso realizzato presentando i risultati raggiunti e lasciando spunto per possibili miglioramenti ed aggiunte.