Entity

Time filter

Source Type

Milano, Italy

Colone F.,University of Rome La Sapienza | Pastina D.,University of Rome La Sapienza | Falcone P.,University of Rome La Sapienza | Falcone P.,Aresys Srl | Lombardo P.,University of Rome La Sapienza
IEEE Transactions on Geoscience and Remote Sensing | Year: 2014

This paper presents an effective signal processing scheme to track moving vehicles and to obtain their cross-range profiles with a passive bistatic radar (PBR) based on the signals of opportunity emitted by a WiFi router. While the target detection using WiFi-based PBR has already been studied by the authors, this paper focuses on the targets moving with a low radial velocity component. These are especially interesting since they might have a reasonable cross-range velocity component, which allows us to apply inverse synthetic aperture radar (ISAR) techniques to provide a high-resolution cross-range profile. A specific problem for these targets is the presence of possibly strong echoes from the stationary background (clutter), which tend to mask their contributions. In such cases, the standard Doppler processing does not help in separating the targets from this clutter. Therefore, appropriate clutter cancellation schemes are applied, and their effectiveness and impact are analyzed both on the tracking and on the ISAR profiling. An appropriate ISAR scheme for cross-range profiling is introduced, tailored for the typical short-range and possibly bistatic surveillance scenarios of the WiFi-based PBR; this scheme comprises the automatic estimation from the data of the target motion components up to a higher order than in usual long-range imaging and their compensation. The reliability of the obtained profiles is also investigated, for both the monostatic and bistatic cases, which is essential both for the vehicle size/characteristics estimation and for the automatic recognition schemes based on vehicle databases. The results obtained using an experimental setup developed and fielded at the University of Rome "La Sapienza," Rome, Italy, show that the considered approach is effective and that the obtained cross-range profiles achieved by ISAR processing with WiFi-based passive radar are quite reliable both in the monostatic and bistatic cases. © 2013 IEEE. Source


Pastina D.,University of Rome La Sapienza | Colone F.,University of Rome La Sapienza | Martelli T.,University of Rome La Sapienza | Falcone P.,Aresys Srl
IEEE Transactions on Vehicular Technology | Year: 2015

In this paper, we examine the potentiality of passive coherent location (PCL) for indoor area monitoring. In particular, we show that Wi-Fi transmissions can be successfully exploited as waveforms of opportunity to perform moving target detection and localization based on the passive radar principle. Moreover, we investigate the advanced capability to obtain high-resolution cross-range profiles of the observed targets via inverse-synthetic-aperture-radar (ISAR) techniques. To these purposes, appropriate processing techniques are introduced to cope with the limitations resulting from the indoor applications such as the strong returns from the stationary scene and the high density of potential targets. The proposed system concept has been tested against both simulated and real data sets. The reported results clearly show that using few receiving channels connected to properly dislocated antennas allows an accurate target localization and tracking. In addition, reliable and stable profiles are obtained for the targets moving in the surveyed scene, which might fruitfully feed a classification stage. This contributes to the demonstration of the effective applicability of the passive radar concept for improving internal and external security of private/public premises. © 1967-2012 IEEE. Source


Villa A.,Aresys Srl | Villa A.,Polytechnic of Milan | Chanussot J.,CNRS GIPSA Laboratory | Benediktsson J.A.,University of Iceland | And 2 more authors.
Pattern Recognition | Year: 2013

The problem of structure detection and unsupervised classification of hyperspectral images with low spatial resolution is addressed in this paper. Hyperspectral imaging is a continuously growing area in remote sensing applications. The wide spectral range, providing a very high spectral resolution, allows the detection and classification surfaces and chemical elements of the observed image. The main problem of hyperspectral images is that the spatial resolution can vary from a few to tens of meters. Many factors, such as imperfect imaging optics, atmospheric scattering, secondary illumination effects and sensor noise cause a degradation of the acquired image quality, making the spatial resolution one of the most expensive and hardest to improve in imaging systems. Due to such a constraint, mixed pixels, e.g., pixels containing mixture of different materials, are quite common in hyperspectral images. In this work, we exploit the rich spectral information of hyperspectral images to deal with the problem. Two methods, based on the concept of spectral unmixing and unsupervised classification, are proposed to obtain thematic maps at a finer spatial scale in a totally unsupervised way. Experiments are carried out on one simulated and two real hyperspectral data sets and clearly show the comparative effectiveness of the proposed method with respect to traditional unsupervised methods, both for classification and detection of spatial structures. © 2012 Elsevier Ltd. Source


Scagliola M.,Aresys Srl | Dinardo S.,Serco | Fornari M.,European Space Agency
International Geoscience and Remote Sensing Symposium (IGARSS) | Year: 2015

One of the main benefits of the along-track processing in Synthetic Aperture Radar altimetry [1] is the speckle reduction that is achieved by averaging all the observations accumulated for a same scattering area. Being the different observations obtained by looking at the same scattering area from different positions of the instrument along the orbit, they result to be modulated by the along-track antenna pattern. By compensating the along-track antenna pattern before averaging, a higher speckle reduction can be achieved. This paper is aimed at presenting this processing method and at evaluating the possible improvement in the sense of the effective number of looks and of the precision of the physical parameters retrieved from the power waveforms. © 2015 IEEE. Source


Fauvel M.,French National Institute for Agricultural Research | Chanussot J.,CNRS GIPSA Laboratory | Benediktsson J.A.,University of Iceland | Villa A.,Aresys Srl | Villa A.,Polytechnic of Milan
Pattern Recognition | Year: 2013

The classification of high dimensional data with kernel methods is considered in this paper. Exploiting the emptiness property of high dimensional spaces, a kernel based on the Mahalanobis distance is proposed. The computation of the Mahalanobis distance requires the inversion of a covariance matrix. In high dimensional spaces, the estimated covariance matrix is ill-conditioned and its inversion is unstable or impossible. Using a parsimonious statistical model, namely the High Dimensional Discriminant Analysis model, the specific signal and noise subspaces are estimated for each considered class making the inverse of the class specific covariance matrix explicit and stable, leading to the definition of a parsimonious Mahalanobis kernel. A SVM based framework is used for selecting the hyperparameters of the parsimonious Mahalanobis kernel by optimizing the so-called radius-margin bound. Experimental results on three high dimensional data sets show that the proposed kernel is suitable for classifying high dimensional data, providing better classification accuracies than the conventional Gaussian kernel. © 2012 Elsevier Ltd. Source

Discover hidden collaborations