Image recognition is an open challenge in computer vision since its early stages. The application of deep neural networks yielded significant improvements towards its solution. Despite their classification abilities, deep networks need datasets with thousands of labelled images and prohibitive computational capabilities to achieve good performance. To address some of these challenges, the CapsNet neural architecture has been recently proposed as a promising machine learning model for image classification based on the idea of capsules. A capsule is a group of neurons whose output represents the presence of features of the same entity. In this paper, we start from the CapsNet architecture to explore and analyse the interaction between the presence of features within certain, similar classes. This is achieved by means of techniques for the features interaction, working on the outputs of two independent capsule-based models. To understand the importance of the interaction between capsules, extensive experiments have been carried out on four challenging dataset. Results show that the exploitation of capsules interaction yields to performance improvements.

The paper proposes a conceptual model inspired to semiotic theories, to be applied to the analysis and design of multimedia. We introduce a meta-model with four levels of semantic aggregation. There results a framework of concepts, relations and processes accounting for the multiple meanings that arise from a multimedia text. We explore the effectiveness of the model by considering four commercial clips by the brand Lancôme. Our analysis confirms that the framework is well suited to the analysis, indexing, design of narrative multimedia.

We address the conceptual modelling of hypermedia regarded as semiotic texts, whose meanings are conceived by a designer, transferred through the artifact and interpreted by users within their context. We outline the communication framework, with the artifact embedding the images of the designer and user. The full model is represented in terms of four interrelated modules: story, discourse, text and social-relational ontologies. The first and second one account for the narrative structures underlying the hypermedia content, which is externalized through sensorial qualities that, in turn, evoke impressions into the user. The model addresses issues that are poorly covered by the description standard MPEG-7. It can be used for analysis, evaluation, indexing of existing hypermedia as well as the design of new ones.

Basic ideas are presented of a multi-national research project to share data about the Roman city of Aquileia employing the Information and Communication Technologies (ICT). A Consortium is proposed to manage the project, adopting the Open-source approach to the software design. The Consortium proposes a common vision of the data which we detail in the paper. There results a shared vocabulary of terms and meanings, as well as standard metadata formats to encode, classify and exchange data from whatever source. A federated system of computer resources realizes and supports the project. We also discuss the results to be realistically expected in short time of a low-cost, joint research effort.

An abstract is not available for this content so a preview has been provided. Please use the Get access link above for information on how to access this content.

The paper proposes a semiotic framework inspired to the Generative Trajectory of Meaning by A.J. Greimas to the design and analysis of hypermedia. The framework is structured into four levels of signification and allows the designer to control the semantic coherence and optimize the communication during the hypermedia development. We investigate its practical feasibility by constructing an instructional hypermedia regarding a collection of ancient mosaics from a roman villa. Our results indicate that the proposed approach opens novel perspectives, and looks promising towards the definition of semiotic methodologies of hypermedia design.

Theia is a software framework for multispectral image analysis. The design is grounded on the object-oriented approach and a model combining parallel computation with selective data processing. Multispectral images from the archaeological site of Aquileia, Italy, have been used as the experimental testbed in order to assess the effectiveness and performance of the system; satisfactory results are reported, and are quite promising towards the use of the framework as a dynamic, interactive interface to real-time data exploration and processing.

The paper proposes a semiotic model inspired to the narrative, structural framework by A.J.Greimas, and applied to the design and analysis of a hypermedia. The framework is structured into four levels of signification and allows for a constructive process of semantic aggregation. We follow such a process in order to explore the conceptual organization of the model. We construct a hypermedia regarding a collection of ancient mosaics from a roman villa. Our results indicate that the proposed approach opens novel perspectives, and looks promising towards the definition of semiotic methodologies of hypermedia design.

Preliminary results of Theia, a software system for multispectral image visualization and analysis are presented. A new approach is adopted, based on modern design techniques and better tuned to the recent advancements in hardware. A careful implementation in the C++ language addresses the issues of time efficiency, openness to personalizations and portability by exploiting the advances of Open Source technologies. Experimental tests on multispectral images have given promising results towards the use of the system as a dynamic, interactive interface to massive data visualization, mining and processing.

Reuse and integration are major steps in the ontology development process, often unavoidable to lower the costs of a new application. We address the two issues with an engineering approach, and provide a comprehensive account of the state of the art. The domain of e-learning is chosen as a test-bed to verify the feasibility of methods. Reuse and integration are shown to be effective in constructing domain representations with larger coverage and enhanced reasoning capabilities, which significantly extend the expressiveness of semantic software. Problems emerge from the lack of explicit representations of the assumptions taken: when reusing an ontology, one implicitly commits to the underlying context. We conclude that a new methodology should be put forward that views the development as a collaborative work within a community of stakeholders.

Data mining techniques, including clustering and classification tasks, for the automatic information extraction from large datasets are increasingly demanded in several scientific fields. Inparticular, in the astrophysical field, large archives and digital sky surveys with dimensions of 1012 bytes currently exist, while in the near future they will reach sizes of the order of 1015. In this work we propose a multidimensional indexing method to effciently query and mine large astrophysical datasets. A novelty detection algorithm, based on the Support Vector Clustering and using density and neighborhood information storedin the index structure, is proposed to find regions of interest in data characterized by isotropic noise. We show an application of this method for the detection of point sources from a gamma-ray photon list.

Cognitive impairment, especially in its early stages, is associated with very mild signs and symptoms that are difficult to detect by clinical and neuropsychological assessment. Advanced imaging analysis techniques applied to magnetic resonance images allow the detection of cerebral structural changes in vivo in mildly affected patients, and might be a useful supporting tool in the early diagnosis and treatment of patients with cognitive impairment. The increasing importance of computer science in cognitive neuroscience has led to the dissemination of a new discipline, neuroinformatics, which is crucial for the introduction of research findings into clinical practice. This review describes some advanced imaging analysis techniques aimed at studying brain structural images and how these techniques might benefit clinical practice through image data sharing and remote analysis in order to increase the accuracy of diagnosis in patients with cognitive impairment.

This correspondence derives a formal link between temporally weighted frame differences, or disturbance fields, which carry limited information suitable for motion detection, and the optic flow (OF), which carries richer information on local image motion. We use this link to derive a novel, simple, near-recursive optic flow algorithm based on a recursive-filter formulation. Most quantities involved are computed recursively, using only data from the current and previous frame. We can limit expensive OF calculations to pixels where motion magnitude is sufficiently high using image differences which the algorithm computes anyway. Experimental results with well-known synthetic, ground-truthed test sequences and standard performance metrics indicate good quantitative performance. Tests with real sequences suggest similar or better performance than a well-known, similar algorithm due to Lucas and Kanade (LK).

Issues of semantic interoperability and service integration for e-government portals are the domain of interest of the present paper. We propose a Conceptual Model for One-Stop e-Government Portals based on the Semantic Web Service technology. We describe our research into building the three basic ontologies and their integration with standard ontologies. The result is a project-independent reusable model. At the same time, we outline a simple methodology for applying the proposed conceptual model into a specific scenario.

We present a web-based system for remotely reconstructing 3D models from 2D images. The system relies on a full static stereo vision procedure, which includes feature extraction; matching; fundamental matrix and roto-translation parameters estimation; image rectification; 3D-point triangulation and poly-hedrization, to produce a 3D geometrical model approximating the input scene. The system architecture, grounded on a conventional client-server scheme, has been designed according to the Model-View-Controller (MVC) paradigm in order to obtain a modular configuration, keeping separate the presentation from the processing and control phases. The resulting, flexible scheme allows the user to fully control the flow of information by means of an interactive interface. In this way, the partial results arising from the numerical steps can be refined by the user, who can modify input parameters, add/eliminate/change output results, thereby actively contributing to an accurate and efficient synthesis of the final model. The system has been realized on an open-source platform; results are presented and discussed.

This paper addresses the problem of computing the sequence of positions in space (inspection script) from which a robot-mounted sensor can perform a given inspection task optimally. We present the design and partial realisation of a sensor planning system, GASP, capable of generating inspection scripts for a variety of tasks, objects and sensors. The treatment of detection uncertainty is discussed with reference to the problem of detecting line features; the generalised sensor is introduced, and novel results in a simulated environment are reported.

We present a new, efficient stereo algorithm addressing robust disparity estimation in the presence of occlusions. The algorithm uses multiple windows and left-right consistency to compute disparity and its associated uncertainty. We demonstrate and discuss performances with both synthetic and real stereo pairs, and show how our results improve on those of closely related techniques for both robustness and efficiency.

We propose a semantically-enhanced architecture to address the issues of interoperability and service integration in e-government web information systems. An architecture for a life event portal based on Semantic Web Services (SWS) is described. The architecture includes loosely-coupled modules organized in three distinct layers: User Interaction, Middleware and Web Services. The Middleware provides the semantic infrastructure for ontologies and SWS. In particular a conceptual model for integrating domain knowledge (Life Event Ontology), application knowledge (E-government Ontology) and service description (Service Ontology) is defined. The model has been applied to a use case scenario in e-government and the results of a system prototype have been reported to demonstrate some relevant features of the proposed approach.