Remote sensing optical images of very high geometrical resolution can provide a precise and detailed representation of the surveyed scene. Thus, the spatial information contained in these images is fundamental for any application requiring the analysis of the image. However, modeling the spatial information is not a trivial task. We addressed this problem by using operators defined in the mathematical morphology framework in order to extract spatial features from the image. In this thesis novel techniques based on mathematical morphology are presented and investigated for the analysis of remote sensing optical images addressing different applications. Attribute Profiles (APs) are proposed as a novel generalization based on attribute filters of the Morphological Profile operator. Attribute filters are connected operators which can process an image by removing flat zones according to a given criterion. They are flexible operators since they can transform an image according to many different attributes (e.g., geometrical, textural and spectral). Furthermore, Extended Attribute Profiles (EAPs), a generalization of APs, are presented for the analysis of hyperspectral images. The EAPs are employed for including spatial features in the thematic classification of hyperspectral images. Two techniques dealing with EAPs and dimensionality reduction transformations are proposed and applied in image classification. In greater detail, one of the techniques is based on Independent Component Analysis and the other one deals with feature extraction techniques. Moreover, a technique based on APs for extracting features for the detection of buildings in a scene is investigated. Approaches that process an image by considering both bright and dark components of a scene are investigated. In particular, the effect of applying attribute filters in an alternating sequential setting is investigated. Furthermore, the concept of Self-Dual Attribute Profile (SDAP) is introduced. SDAPs are APs built on an inclusion tree instead of a min- and max-tree, providing an operator that performs a multilevel filtering of both the bright and dark components of an image. Techniques developed for applications different from image classification are also considered. In greater detail, a general approach for image simplification based on attribute filters is proposed. Finally, two change detection techniques are developed. The experimental analysis performed with the novel techniques developed in this thesis demonstrates an improvement in terms of accuracies in different fields of application when compared to other state of the art methods.

In today’s digital age, the enhancement of the hardware technology has set new horizons on the computer science universe, asking new questions, proposing new solutions and re-opening some branches that have been temporary closed due to the overwhelming computational complexity. In this sense many algorithms have been proposed but they have never been successfully applied in practice up to now. In this work we will tackle the issues related to the detection and the localization of an interaction conducted by humans. We will begin analysing group interactions then moving to dyadic interactions and then elevate our considerations to the real world scenario. We will propose new challenging datasets, introducing new important tasks and suggesting some possible solutions.

The thesis focuses on predicting tolerance effects on the radiation pattern of reflectarray antennas through Interval Analysis. In fact, the uncertainty on the actual size of all parameters under fabrication tolerates such as element dimensions and dielectric properties, are modeled with interval values. Afterwards, the rules of Interval Arithmetic are exploited to compute the bounds of deviation in the resonance frequency of each element, the phase response of the element and the radiated power pattern. Due to the redundancy problems of using Interval Cartesian (IA−CS) for complex structure, the interval bounds are overestimated and the reasons are the Dependency and Wrapping effects of using interval analysis for complex structures. Different techniques are proposed and assessed in order to eliminate the dependency effect such as reformulating the interval function and the Enumerative interval analysis. Moreover, the Minkowski sum approach is used to eliminate the wrapping effect. In numerical validation, a set of representative results, show the power bounds computations with Interval Cartesian method (IA − CS), a modified Interval Cartesian method (IA − CS*), Interval Enumerative method (IA − ENUM) and Interval Enumerative Minkowski method ( IA − ENUM − MS) and a comparative study is reported in order to assess the effectiveness of the proposed approach (IA − ENUM − MS) with respect to the other methods. Furthermore, different tolerances in patch width,length, substrate thickness and dielectric permittivity are considered which shows that the higher uncertainty produces the larger deviation of the pattern bounds and the larger deviation include the smaller deviation and the nominal one. To validate the inclusion properties of the interval bounds, the results are compared with Monte Carlo simulation results. Then, a numerical study is devoted to analyse the dependency of the degradation of the pattern features to steering angle and the bandwidth. Finally, the effect of feed displacement errors on the power pattern of reflecttarray antennas is considered with Interval Enumerative Minkowski method. The maximal deviations from the nominal power pattern (error free) and its features are analysed for several reflectarray structures with different focal-length-to-diameter ratios to prove the effectiveness of the proposed method.

Deep learning, a branch of machine learning, has been gaining ground in many research fields as well as practical applications. Such ongoing boom can be traced back mainly to the availability and the affordability of potential processing facilities, which were not widely accessible than just a decade ago for instance. Although it has demonstrated cutting-edge performance widely in computer vision, and particularly in object recognition and detection, deep learning is yet to find its way into other research areas. Furthermore, the performance of deep learning models has a strong dependency on the way in which these latter are designed/tailored to the problem at hand. This, thereby, raises not only precision concerns but also processing overheads. The success and applicability of a deep learning system relies jointly on both components. In this dissertation, we present innovative deep learning schemes, with application to interesting though less-addressed topics. In this respect, the first covered topic is rough scene description for visually impaired individuals, whose idea is to list the objects that likely exist in an image that is grabbed by a visually impaired person, To this end, we proceed by extracting several features from the respective query image in order to capture the textural as well as the chromatic cues therein. Further, in order to improve the representativeness of the extracted features, we reinforce them with a feature learning stage by means of an autoencoder model. This latter is topped with a logistic regression layer in order to detect the presence of objects if any. In a second topic, we suggest to exploit the same model, i.e., autoencoder in the context of cloud removal in remote sensing images. Briefly, the model is learned on a cloud-free image pertaining to a certain geographical area, and applied afterwards on another cloud-contaminated image, acquired at a different time instant, of the same area. Two reconstruction strategies are proposed, namely pixel-based and patch-based reconstructions. From the earlier two topics, we quantitatively demonstrate that autoencoders can play a pivotal role in terms of both (i) feature learning and (ii) reconstruction and mapping of sequential data. Convolutional Neural Network (CNN) is arguably the most utilized model by the computer vision community, which is reasonable thanks to its remarkable performance in object and scene recognition, with respect to traditional hand-crafted features. Nevertheless, it is evident that CNN naturally is availed in its two-dimensional version. This raises questions on its applicability to unidimensional data. Thus, a third contribution of this thesis is devoted to the design of a unidimensional architecture of the CNN, which is applied to spectroscopic data. In other terms, CNN is tailored for feature extraction from one-dimensional chemometric data, whilst the extracted features are fed into advanced regression methods to estimate underlying chemical component concentrations. Experimental findings suggest that, similarly to 2D CNNs, unidimensional CNNs are also prone to impose themselves with respect to traditional methods. The last contribution of this dissertation is to develop new method to estimate the connection weights of the CNNs. It is based on training an SVM for each kernel of the CNN. Such method has the advantage of being fast and adequate for applications that characterized by small datasets.

Time synchronization is of primary importance for the operation of wireless sensor networks (WSN): time measurements, coordinated actions and event ordering require common time on WSN nodes. Due to intrinsic energy limitations of wireless networks there is a need for new energy-efficient time synchronization solutions, different from the ones that have been developed for wired networks. In this work we investigated the trade-offs between time synchronization accuracy and energy saving in WSN. On the basis of that study we developed a power-efficient adaptive time synchronization strategy, that achieves a target synchronization accuracy at the expense of a negligible overhead. Also, we studied the energy benefits of periodic time synchronization in WSN employing synchronous wakeup schemes, and developed an algorithm that finds the optimal synchronization period to save energy. The proposed research improves state-of-the-art by exploring new ways to save energy while assuring high flexibility and reliable operation of WSN.

The widespread use of digital cameras, as well as the increasing popularity of online photo sharing has led to the proliferation of networked photo collections. Handling such a huge amount of media, without imposing complex and time consuming archiving procedures, is highly desirable and poses a number of interesting research challenges to the media community. In particular, the definition of suitable content based indexing and retrieval methodologies is attracting the effort of a large number of researchers worldwide, who proposed various tools for automatic content organization, retrieval, search, annotation and summarization. In this thesis, we will present and discuss three different approaches for content-and-context based retrieval. The main focus will be put on personal photo albums, which can be considered one of the most challenging application domains in this field, due to the largely unstructured and variable nature of the datasets. The methodologies that we will describe can be summarized into the following three points: i. Stochastic approaches to exploit the user interaction in query-by-example photos retrieval. Understanding the subjective meaning of a visual query, by converting it into numerical parameters that can be extracted and compared by a computer, is the paramount challenge in the field of intelligent image retrieval, also referred to as the “semantic gap†problem. An innovative approach is proposed that combines a relevance feedback process with a stochastic optimization engine, as a way to grasp user's semantics through optimized iterative learning providing on one side a better exploration of the search space, and on the other side avoiding stagnation in local minima during the retrieval. ii. Unsupervised event collection, segmentation and summarization. The need for automatic tools able to extract salient moments and provide automatic summary of large photo galleries is becoming more and more important due to the exponential growth in the use of digital media for recording personal, familiar or social life events. The multi-modal event segmentation algorithm faces the summarization problem in an holistic way, making it possible to exploit the whole available information in a fully unsupervised way. The proposed technique aims at providing such a tool, with the specific goal of reducing the need of complex parameter settings and letting the system be widely useful for as many situations as possible. iii. Content-based synchronization of multiple galleries related to the same event. The large spread of photo cameras makes it quite common that an event is acquired through different devices, conveying different subjects and perspectives of the same happening. Automatic tools are more and more used to support the users in organizing such archives, and it is largely accepted that time information is crucial to this purpose. Unfortunately time-stamps may be affected by erroneous or imprecise setting of the camera clock. The synchronization algorithm presented is the first that uses the content of pictures to estimate the mutual delays among different cameras, thus achieving an a-posteriori synchronization of various photo collections referring to the same event.

The study, development, and analysis of innovative inversion techniques for the detection and imaging of buried objects is addressed in this thesis. The proposed methodologies are based on the use of microwave radiations and radar techniques for subsurface prospecting, such as, for example, the Ground Penetrating Radar (GPR). More precisely, the reconstruction of shallow buried objects is firstly addressed by an electromagnetic inverse scattering method based on the integration of the inexact Newton (IN) method with an iterative multiscaling approach. The performances of such an inversion approach are analyzed both when considering the use of a second-order Born approximation (SOBA) and when exploiting the full set of non-linear equations governing the scattering phenomena for the buried scenario. The presented methodologies are particularly suitable for applications such as demining (e.g., for the detection of unexploded ordnances, UXOs, and improvised explosive devices, IEDs), for civil engineering applications (e.g., for the investigation of possible structural damages, voids, cracks or water infiltrations in walls, pillars, bridges) as well as for biomedical imaging (e.g., for early cancer detection).

In the last decade the interest in radar systems for the exploration of planetary bodies and for Earth Observation (EO) from orbit increased considerably. In this context, the main goal of this thesis is to present novel methods for the automatic analysis of planetary radar sounder (RS) signals and very high resolution (VHR) synthetic aperture radar (SAR) images acquired on the Earth. Both planetary RSs and VHR SAR systems are instruments based on relatively recent technology which make it possible to acquire from orbit new types of data that before were available only in limited areas from airborne acquisitions. The use of orbiting platforms allows the acquisition of a huge amount of data on large areas. This calls for the development of effective and automatic methods for the extraction of information tuned on the characteristics of these new systems. The work has been organized in two parts. The first part is focused on the automatic analysis of data acquired by planetary RSs. RS signals are currently mostly analyzed by means of manual investigations and the topic of automatic analysis of such data has been only marginally addressed in the literature. In this thesis we provide three main novel contributions to the state of the art on this topic. First, we present a theoretical and empirical statistical study of the properties of RS signals. Such a study drives the development of two novel automatic methods for the generation of subsurface feature maps and for the detection of basal returns. The second contribution is a method for the extraction of subsurface layering in icy environments, which is capable to detect linear features with sub-pixel accuracy. Moreover, measures for the analysis of the properties of the detected layers are proposed. Finally, the third contribution is a technique for the detection of surface clutter returns in radargrams. The proposed method is based on the automatic matching between real and clutter data generated according to a simulator developed in this thesis. The second part of this dissertation is devoted to the analysis of VHR SAR images, with special focus on urban areas. New VHR SAR sensors allow the analysis of such areas at building level from space. This is a relatively recent topic, which is especially relevant for crisis management and damage assessment. In this context, we describe in detail an empirical and theoretical study carried out on the relation between the double-bounce effect of buildings and their orientation angle. Then, a novel approach to the automatic detection and reconstruction of building radar footprints from VHR SAR images is pre-sented. Unlike most of the methods presented in the literature, the developed method can extract and reconstruct building radar footprints from single VHR SAR images. The technique is based on the detection and combination of primitive features in the image, and introduces the concept of semantic meaning of the primitives. Qualitative and quantitative experimental results obtained on real planetary RS and spaceborne VHR SAR data confirm the effectiveness of the proposed methods.

Future mobile broadband communications working over wireless channels are required to provide high performance services in terms of speed, capacity, and quality. A key issue to be considered is the design of multi-standard and multi-modal ad-hoc network architectures, capable of self-configuring in an adaptive and optimal way with respect to channel conditions and traffic load. In the context of 4G-wireless communications, the implementation of efficient baseband receivers characterized by affordable computational load is a crucial point in the development of transmission systems exploiting diversity in different domains. This thesis proposes some novel multi-user detection techniques based on different criterions (i.e., MMSE, ML, and MBER) particularly suited for multi-carrier CDMA systems, both in the single- and multi-antenna cases. Moreover, it considers the use of evolutionary strategies (such as GA and PSO) to channel estimation purposes in MIMO multicarrier scenarios. Simulation results evidenced that the proposed PHY-layer optimization techniques always outperform state of the art schemes by spending an affordable computational burden. Particular attention has been used on the software implementation of the formulated algorithms, in order to obtain a modular software architecture that can be used in an adaptive and optimized reconfigurable scenario.

The study, design and development of a monitoring system for wildlife road crossing problem is addressed in this thesis. Collisions between fauna and vehicles is a relevant issue in several mountain and rural regions and a valuable low-cost solution has not yet been identified. In particular, the proposed system is composed by a network of sensors installed along road margins, in order to detect wildlife events, (e.g., approaching, leaving or crossing the road), thus to promptly warn the incoming drivers. The sensor nodes communicate wirelessly among the network thus collecting the sensed information in a control unit for data storage, processing and statistics. The detection process is performed by the wireless nodes, which are equipped with low-cost Doppler radars for real-time identification of wildlife movements. In detail, different technologies valuable for solving the problem and related off-the-shelf solutions have been investigated and properly tested in order to validate their actual performance considering the specific problem scenario. A final classification based on specific parameters has allowed identifying the Doppler radar system as the better low-cost technology for contributing to the problem objective. The performance of the proposed system has also been investigated in a real scenario, which has been identified to be the actual pilot site for the monitoring system. This confirms the system capability of movements detection in the road proximity, thus defining a security area along it, where all occurring events may be identified.