LISTIC

Chambéry, France
Chambéry, France
SEARCH FILTERS
Time filter
Source Type

Breloy A.,Paris West University Nanterre La Défense | Sun Y.,Purdue University | Babu P.,Indian Institute of Technology Delhi | Ginolhac G.,LISTIC | Palomar D.P.,DECE
Conference Record - Asilomar Conference on Signals, Systems and Computers | Year: 2016

In this paper, we consider the problem of robustly estimating a structured covariance matrix (CM). Specifically, we focus on CM structures that involve Kronecker products of low rank matrices, which often arise in the context of array processing (e.g. in MIMO-Radar, COLD array, and STAP). To tackle this problem, we derive a new Constrained Tyler's Estimators (CTE), which is defined as the minimizer of the cost function associated to Tyler's estimator under Kronecker structural constraint. Algorithms to compute these new CTEs are derived based on the Majorization-Minimization algorithmic framework. © 2016 IEEE.


Destercke S.,Compiègne University of Technology | Antoine V.,LISTIC
Advances in Intelligent Systems and Computing | Year: 2013

When working with sets of probabilities, basic information fusion operators quickly reach their limits: intersection becomes empty, while union results in a poorly informative model. An attractive means to overcome these limitations is to use maximal coherent subsets (MCS). However, identifying the maximal coherent subsets is generally NP-hard. Previous proposals advocating the use of MCS to merge probability sets have not provided efficient ways to perform this task. In this paper, we propose an efficient approach to do such a merging between imprecise probability masses, a popular model of probability sets, and test it on an ensemble classification problem. © 2013 Springer-Verlag.


Yan Y.,LISTIC | Doin M.-P.,CNRS ENS Geology Laboratory | Lopez-Quiroz P.,National Autonomous University of Mexico | Tupin F.,Telecom ParisTech | And 3 more authors.
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing | Year: 2012

In multi-temporal InSAR processing, both the Permanent Scatterer (PS) and Small BAseline Subset (SBAS) approaches are optimized to obtain ground displacement rates with a nominal accuracy of millimeters per year. In this paper, we investigate how applying both approaches to Mexico City subsidence validates the InSAR time series results and brings complementary information to the subsidence pattern. We apply the PS approach (Gamma-IPTA chain) and an ad-hoc SBAS approach on 38 ENVISAT images from November 2002 to March 2007 to map the Mexico City subsidence. The subsidence rate maps obtained by both approaches are compared quantitatively and analyzed at different steps of the PS processing. The inter-comparison is done separately for low-pass (LP) and high-pass (HP) filtered difference maps to take the complementarity of both approaches at different scales into account. The inter-comparison shows that the differential subsidence map obtained by the SBAS approach describes the local features associated with urban constructions and infrastructures, while the PS approach quantitatively characterizes the motion of individual targets. The latter information, once related to the type of building foundations, should be essential to quantify the relative importance of surface loads, surface drying and drying due to aquifer over-exploitation, in subsoil compaction. © 2012 IEEE.


Berrah L.,LISTIC | Montmain J.,Ecole des Mines d'Ales | Mauris G.,LISTIC | Cliville V.,LISTIC
International Journal of Data Analysis Techniques and Strategies | Year: 2011

The major industrial control purpose is the reaching of the expected performances. In this sense, improvement processes are continuously carried out in order to define the right actions with regard to the objectives achievement. Thus, in order to better monitor the performance continuous improvement process, we consider a quantitative model for performance assessment. The industrial performance being multi-criteria, the proposed model is thus based on the one hand, on the MACBETH method to express quantitatively elementary performances from qualitative expert pair-wise comparisons and, on the other hand, on the Choquet integral to express the overall performance according to subordination and transverse interactions between the elementary performances. Then, the main focus concerns the decision-maker's requirements for optimising the improvement of the overall performance versus the allocated resources. In this view, we propose useful pieces of information first for diagnosis, then for overall performance improvement optimisation versus the costs of elementary performance improvements. Finally, the proposed approach is applied to an industrial case looking for optimising the improvement of the lean objective satisfaction related to the throughput time of hydraulic component manufacturing. Copyright © 2011 Inderscience Enterprises Ltd.


Benoit A.,LISTIC | Caplier A.,CNRS GIPSA Laboratory
Computer Vision and Image Understanding | Year: 2010

This paper proposes to demonstrate the advantages of using certain properties of the human visual system in order to develop a set of fusion algorithms for automatic analysis and interpretation of global and local facial motions. The proposed fusion algorithms rely on information coming from human vision models such as human retina and primary visual cortex previously developed at Gipsa-lab. Starting from a set of low level bio-inspired modules (static and moving contour detector, motion event detector and spectrum analyser) which are very efficient for video data pre-processing, it is shown how to organize them together in order to achieve reliable face motion interpretation. In particular, algorithms for global head motion analysis such as head nods, for local eye motion analysis such as blinking, for local mouth motion analysis such as speech lip motion and yawning and for open/close mouth/eye state detection are proposed and their performances are assessed. Thanks to the use of human vision model pre-processing which decorrelates visual information in a reliable manner, fusion algorithms are simplified and remain robust against traditional video acquisition problems (light changes, object detection failure, etc.). © 2010 Elsevier Inc.


Brigui F.,Nanyang Technological University | Thirion-Lefevre L.,Supelec | Ginolhac G.,LISTIC | Forster P.,Ecole Normale Superieure de Cachan
IEEE Transactions on Geoscience and Remote Sensing | Year: 2014

We develop a new synthetic aperture radar (SAR) algorithm based on physical models for the detection of a man-made target (MMT) embedded in strong interferences (trunks of a forest). These physical models for the MMT and the interferences are integrated in low-rank subspaces and are based on scattering and polarimetric properties. Several images, called subspace SAR images, can be generated and combined considering these subspace models. We then propose a new approach for target detection and interference reduction based on the combination of SAR subspace images. We show that our SAR algorithm outperforms the classical SAR imagery algorithm on both simulated data and real data in the context of foliage penetration detection. © 2013 IEEE.


Combernoux A.,Supelec | Pascal F.,Supelec | Ginolhac G.,LISTIC | Lesturgie M.,Supelec
ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings | Year: 2015

The paper addresses the problem of approximating the detector distribution used in target detection embedded in a disturbance composed of a low rank Gaussian noise and a white Gaussian noise. In this context, it is interesting to use an adaptive version of the Low Rank Normalized Matched Filter (LR-ANMF) detector, which is a function of the estimated projector onto the low rank noise subspace. We will show that the traditional approximation of the LR-ANMF detector distribution is not always the better one. In this paper, we propose to perform its limits when the number of secondary data K and the data dimension m both tend to infinity at the same rate m/K → c element 2 (0;∞). Then, we give the theoretical distributions of these limits in the large dimensional regime and approximate the LR-ANMF detector distribution by them. The comparison of empirical and theoretical distributions on a jamming application shows the interest of our approach. © 2015 IEEE.


Combernoux A.,Supelec | Pascal F.,Supelec | Ginolhac G.,LISTIC | Lesturgie M.,Supelec
European Signal Processing Conference | Year: 2013

The paper addresses the problem of target detection embedded in a disturbance composed of a low rank Gaussian clutter and a white Gaussian noise. In this context, it is interesting to use an adaptive version of the Low Rank Normalized Matched Filter detector, denoted LR-ANMF, which is a function of the estimation of the projector onto the clutter subspace. In this paper, we show that the LR-ANMF detector based on the sample covariance matrix is consistent when the number of secondary data K tends to infinity for a fixed data dimension m but not consistent when m and K both tend to infinity at the same rate, i.e. m/K → c (0, ∞). Using the results of random matrix theory, we then propose a new version of the LR-ANMF which is consistent in both cases. The application of our new detector on STAP (Space Time Adaptive Processing) data shows the interest of our approach. © 2013 EURASIP.


Brigui F.,Nanyang Technological University | Ginolhac G.,LISTIC | Thirion-Lefevre L.,Supelec | Forster P.,Ecole Normale Superieure de Cachan
IEEE Transactions on Aerospace and Electronic Systems | Year: 2014

We have developed a new synthetic aperture radar (SAR) algorithm based on physical models for the detection of a man-made target (MMT) embedded in strong clutter (trunks in a forest). The physical models for the MMT and the clutter are represented by low-rank subspaces and are based on scattering and polarimetric properties. Our SAR algorithm applies the oblique projection of the received signal along the clutter subspace onto the target subspace. We compute its statistical performance in terms of probabilities of detection and false alarms. The performances of the proposed SAR algorithm are improved compared to those obtained with existing SAR algorithms: the MMT detection is greatly improved, and the clutter is rejected. We also studied the robustness of our SAR algorithm to interference modeling errors. Results on real foliage penetration data showed the usefulness of this approach. © 2014 IEEE.


Ionescu B.,Polytechnic University of Bucharest | Lambert P.,LISTIC
UPB Scientific Bulletin, Series C: Electrical Engineering | Year: 2013

We address a particular case of video genre classification, namely the classification of animated movies. We propose two categories of content descriptors: temporal and color based, which are adapted to this particu- lar task. Temporal descriptors, like rhythm or action, are quantifying the perception of the action content at different levels. Color descriptors are de- termined using color perception which is quantified in terms of statistics of color distribution, elementary hues, color properties and color relationship. Experimental tests conducted on more than 159 hours of video footage and various classification schemes show the efficiency of this approach. Despite the high diversity of the video material, the proposed descriptors are able to provide an average global correct classification up to 92.7%.

Loading LISTIC collaborators
Loading LISTIC collaborators