Laboratoire Mathématiques, Image et Applications

Background modeling is often used in the context of moving objects detection from static cameras. Numerous methods have been developed over the recent years and the most used are the statistical ones. The purpose of this chapter is to provide a recent survey of these different statistical methods. For this, we have classified them in term of generation following the years of publication and the statistical tools used. We then focus on the first generation methods: Single Gaussian, Mixture of Gaussians, Kernel Density Estimation and Subspace Learning using PCA. These original methods are reminded and then we have classified their different improvements in term of strategies. After analyzing the strategies and identifying their limitations, we conclude with several promising directions for future research.

Human pose estimation refers to the estimation of the location of body parts and how they are connected in an image. Human pose estimation from monocular images has wide applications (e.g., image indexing). Several surveys on human pose estimation can be found in the literature, but they focus on a certain category; for example, model-based approaches or human motion analysis, etc. As far as we know, an overall review of this problem domain has yet to be provided. Furthermore, recent advancements based on deep learning have brought novel algorithms for this problem. In this paper, a comprehensive survey of human pose estimation from monocular images is carried out including milestone works and recent advancements. Based on one standard pipeline for the solution of computer vision problems, this survey splits the problem into several modules: feature extraction and description, human body models, and modeling methods. Problem modeling methods are approached based on two means of categorization in this survey. One way to categorize includes top-down and bottom-up methods, and another way includes generative and discriminative methods. Considering the fact that one direct application of human pose estimation is to provide initialization for automatic video surveillance, there are additional sections for motion-related methods in all modules: motion features, motion models, and motion-based methods. Finally, the paper also collects 26 publicly available data sets for validation and provides error measurement methods that are frequently used.

Abstract This paper deals with the design of modelling tools suitable for investigating the consequences of alternative policies on the dynamics of resources and fisheries, such as the evaluation of marine protected areas (MPA). We first review the numerous models that have been developed for this purpose, and compare them from several standpoints: population modelling, exploitation modelling and management measure modelling. We then present a generic fisheries simulation model, Integration of Spatial Information for FISHeries simulation (ISIS‐Fish). This spatially explicit model allows quantitative policy screening for fisheries with mixed‐species harvests. It may be used to investigate the effects of combined management scenarios including a variety of policies: total allowable catch (TAC), licenses, gear restrictions, MPA, etc. Fisher's response to management may be accounted for by means of decision rules conditioned on population and exploitation parameters. An application to a simple example illustrates the relevance of this kind of tool for policy screening, particularly in the case of mixed fisheries. Finally, the reviewed models and ISIS‐Fish are discussed and confronted in the light of the underlying assumptions and model objectives. In the light of this discussion, we identify desirable features for fisheries simulation models aimed at policy evaluation, and particularly MPA evaluation.

International audience

The analysis and understanding of video sequences is currently quite an active research field. Many applications such as video surveillance, optical motion capture or those of multimedia need to first be able to detect the objects moving in a scene filmed by a static camera. This requires the basic operation that consists of separating the moving objects called from the static information called background. Many background subtraction methods have been developed (Bouwmans et al. (2010); Bouwmans et al. (2008)). A recent survey (Bouwmans (2009)) shows that subspace learning models are well suited for background subtraction. Principal Component Analysis (PCA) has been used to model the background by significantly reducing the data's dimension. To perform PCA, different Robust Principal Components Analysis (RPCA) models have been recently developed in the literature. The background sequence is then modeled by a low rank subspace that can gradually change over time, while the moving foreground objects constitute the correlated sparse outliers. However, authors compare their algorithm only with the PCA (Oliver et al. (1999)) or another RPCA model. Furthermore, the evaluation is not made with the datasets and the measures currently used in the field of background subtraction. Considering all of this, we propose to evaluate RPCA models in the field of video-surveillance. Contributions of this chapter can be summarized as follows: 1) A survey regarding robust principal component analysis and 2) An evaluation and comparison on different video surveillance datasets

Infrared imaging is a crucial technique in a multitude of applications, including night vision, autonomous vehicle navigation, optical tomography, and food quality control. Conventional infrared imaging technologies, however, require the use of materials such as narrow bandgap semiconductors, which are sensitive to thermal noise and often require cryogenic cooling. We demonstrate a compact all-optical alternative to perform infrared imaging in a metasurface composed of GaAs semiconductor nanoantennas, using a nonlinear wave-mixing process. We experimentally show the upconversion of short-wave infrared wavelengths via the coherent parametric process of sum-frequency generation. In this process, an infrared image of a target is mixed inside the metasurface with a strong pump beam, translating the image from the infrared to the visible in a nanoscale ultrathin imaging device. Our results open up new opportunities for the development of compact infrared imaging devices with applications in infrared vision and life sciences.

Let $q\geq 2$ be a positive integer, $B$ be a fractional Brownian motion with Hurst index $H\in$, $Z$ be an Hermite random variable of index $q$, and $H_q$ denote the $q$th Hermite polynomial. For any $n\geq 1$, set $V_n=\sum_{k=0}^{n-1} H_q(B_{k+1}-B_k)$. The aim of the current paper is to derive, in the case when the Hurst index verifies $H<1-1/(2q)$, an upper bound for the total variation distance between the laws $\mathscr{L}(Z_n)$ and $\mathscr{L}(Z)$, where $Z_n$ stands for the correct renormalization of $V_n$ which converges in distribution towards $Z$. Our results should be compared with those obtained recently by Nourdin and Peccati (2007) in the case where $H<1-1/(2q)$, corresponding to the case where one has normal approximation.

Lidar waveforms are 1-D signals representing a train of echoes caused by reflections at different targets. Modeling these echoes with the appropriate parametric function is useful to retrieve information about the physical characteristics of the targets. This paper presents a new probabilistic model based upon a marked point process which reconstructs the echoes from recorded discrete waveforms as a sequence of parametric curves. Such an approach allows to fit each mode of a waveform with the most suitable function and to deal with both, symmetric and asymmetric, echoes. The model takes into account a data term, which measures the coherence between the models and the waveforms, and a regularization term, which introduces prior knowledge on the reconstructed signal. The exploration of the associated configuration space is performed by a reversible jump Markov chain Monte Carlo (RJMCMC) sampler coupled with simulated annealing. Experiments with different kinds of lidar signals, especially from urban scenes, show the high potential of the proposed approach. To further demonstrate the advantages of the suggested method, actual laser scans are classified and the results are reported.

In this paper, we are interested in the study of the asymptotic behavior of a generalization of the Cahn-Hilliard equation with a proliferation term andendowed with Neumann boundary conditions. Such a model has, in particular,applications in biology. We show that either the average of the local density of cells is bounded, in which case we have a global in timesolution, or the solution blows up in finite time.We further prove that the relevant, from a biological point of view, solutions converge to $1$ as time goes to infinity.We finally give some numerical simulations which confirm the theoretical results.

Foreground detection is the first step in video surveillance system to detect moving objects. Principal Components Analysis (PCA) shows a nice framework to separate moving objects from the background but without a mechanism of robust analysis, the moving objects may be absorbed into the background model. This drawback can be solved by recent researches on Robust Principal Component Analysis (RPCA). The background sequence is then modeled by a low rank subspace that can gradually change over time, while the moving foreground objects constitute the correlated sparse outliers. In this paper, we propose to use a RPCA method based on low-rank and block-sparse matrix decomposition to achieve foreground detection. This decomposition enforces the low-rankness of the background and the block-sparsity aspect of the foreground. Experimental results on different datasets show the pertinence of the proposed approach.

In this paper, we study quantitative as well as qualitative properties of Fork-Join Queuing Networks with Blocking (FJQN/Bs). Specifically, we prove results regarding the equivalence of the behavior of a FJQN/B and that of its duals and a strongly connected marked graph. In addition, we obtain general conditions that must be satisfied by the service times to guarantee the existence of a long-term throughput and its independence on the initial configuration. We also establish conditions under which the reverse of a FJQN/B has the same throughput as the original network. By combining the equivalence result for duals and the reversibility result, we establish a symmetry property for the throughput of a FJQN/B. Last, we establish that the throughput is a concave function of the buffer sizes and the initial marking, provided that the service times are mutually independent random variables belonging to the class of PERT distributions that includes the Erlang distributions. This last result coupled with the symmetry property can be used to identify the initial configuration that maximizes the long-term throughput in closed series-parallel networks.

Hysteresis does not vanish in materials essentially free from defects and domain wall pinning. It arises from the existence of many geometrically different domain structures separated by intrinsic energy barriers. Although generally metastable, these states can nevertheless be explored by suitable excursion in the manifold of domain structures, along which irreducible irreversibility is present. This idea is illustrated by experiments performed on garnet layers with uniaxial anisotropy (bubble material). Cooling the layer from T&amp;gt;Tc in a field H normal to it results in various states: parallel stripes (H=0), bubbles of increasing diameters and mixtures of bubbles and stripes (H&amp;lt;52 Oe), and mazes of convoluted stripes with defects (forks, dead ends). Cycling the applied field at constant temperature also causes evolution of the structure. For example, starting from parallel stripes the structure evolves toward a maze with intermediate steps corresponding to stripe folding, and then there is the onset of defects (forks, dead ends). We classify these various structures in a kind of phase diagram, depending on the following parameters: current field H, maximum field Hmax previously applied, and memory temperature Tm at which the domain structure is nucleated. The relations and irreversible paths between various structures are precised. We also discuss mechanisms for topological evolution linked with changes in the equlibrium period of the structure. Irreversibility arises from asymmetric processes of domain collapse and nucleation. The first stages in the processes leading from bubbles to stripes, or from parallel stripes to mazes, can be analyzed respectively in terms of elliptical and folding instabilities, the latter being similar to the undulation instability of smectic A liquid crystals.

The authors present a comparison of fuzzy control algorithms with those of conventional control approaches (PID, optimal) in the case of a linear dynamic process control. They first give an overview of the different controllers used and examine the sensitivity of the fuzzy logic controller to design parameters. A comparative study between three types of controllers with regard to performance and robustness is then presented. It is concluded that the main interest of the fuzzy controller lies in its easy implementation. The expertise and experience of the human operator make it possible to elaborate a set of rules to control the system without requiring the computation of any mathematical model. For simple systems, a real-time adaptation is eased by using quantizations and look-up tables, which result in very short development and perfecting times. A fuzzy controlled system shows good results in terms of response time and precision. Moreover, such a controller allows a good robustness mainly with respect to modifications of process parameters as well as various disturbances on the control.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Abstract Ventifacts, rocks abraded by wind‐borne particles, are found in Gale Crater, Mars. In the eastward drive from “Bradbury Landing” to “Rocknest,” they account for about half of the float and outcrop seen by Curiosity's cameras. Many are faceted and exhibit abrasion textures found at a range of scales, from submillimeter lineations to centimeter‐scale facets, scallops, flutes, and grooves. The drive path geometry in the first 100 sols of the mission emphasized the identification of abrasion facets and textures formed by westerly flow. This upwind direction is inconsistent with predictions based on models and the orientation of regional dunes, suggesting that these ventifact features formed from very rare high‐speed winds. The absence of active sand and evidence for deflation in the area indicates that most of the ventifacts are fossil features experiencing little abrasion today.

For specific applications in the field of high gradient magnetic separation of biomaterials, magnetic nanoparticle clusters of controlled size and high magnetic moment in an external magnetic field are of particular interest. We report the synthesis and characterization of magnetic microgels designed for magnetic separation purposes, as well as the separation efficiency of the obtained microgel particles. High magnetization magnetic microgels with superparamagnetic behaviour were obtained in a two-step synthesis procedure by a miniemulsion technique using highly stable ferrofluid on a volatile nonpolar carrier. Spherical clusters of closely packed hydrophobic oleic acid-coated magnetite nanoparticles were coated with cross linked polymer shells of polyacrylic acid, poly-N-isopropylacrylamide, and poly-3-acrylamidopropyl trimethylammonium chloride. The morphology, size distribution, chemical surface composition, and magnetic properties of the magnetic microgels were determined using transmission electron microscopy, X-ray photoelectron spectroscopy, and vibrating sample magnetometry. Magnetically induced phase condensation in aqueous suspensions of magnetic microgels was investigated by optical microscopy and static light scattering. The condensed phase consists of elongated oblong structures oriented in the direction of the external magnetic field and may grow up to several microns in thickness and tens or even hundreds of microns in length. The dependence of phase condensation magnetic supersaturation on the magnetic field intensity was determined. The experiments using high gradient magnetic separation show high values of separation efficiency (99.9-99.97%) for the magnetic microgels.

We apply the stochastic approximation method to construct a large class of recursive kernel estimators of a distribution function. We study the properties of these estimators and compare them with Nadaraya’s distribution estimator. It turns out that, with an adequate choice of the stepsize of the proposed algorithm, the MWISE (Mean Weighted Integrated Squared Error) of the proposed estimator is smaller than that of Nadaraya’s estimator. We corroborate these theoretical results by simulations and a real dataset.

Abstract Nyamulagira, located in the east of the Democratic Republic of Congo on the western branch of the East African rift, is Africa’s most active volcano, with an average of one eruption every 3 years since 1938. Owing to the socio-economical context of that region, the volcano lacks ground-based geodetic measurements but has been monitored by interferometric synthetic aperture radar (InSAR) since 1996. A combination of 3D Mixed Boundary Element Method and inverse modelling, taking into account topography and source interactions, is used to interpret InSAR ground displacements associated with eruptive activity in 1996, 2002, 2004, 2006 and 2010. These eruptions can be fitted by models incorporating dyke intrusions, and some (namely the 2006 and 2010 eruptions) require a magma reservoir beneath the summit caldera. We investigate inter-eruptive deformation with a multi-temporal InSAR approach. We propose the following magma plumbing system at Nyamulagira by integrating numerical deformation models with other available data: a deep reservoir ( c . 25 km depth) feeds a shallower reservoir ( c . 4 km depth); proximal eruptions are fed from the shallow reservoir through dykes while distal eruptions can be fed directly from the deep reservoir. A dyke-like conduit is also present beneath the upper southeastern flank of Nyamulagira.

Addresses the problem of a temporal signature conservation, arising in multimedia communications. Specifically, the authors are interested in the problem of the flow reconstitution and variable end-to-end delay compensation at the receiver end-system. Requirements and algorithm associated with this design are described. The proposed algorithm is quickly understood, it may exploit either the globally synchronized or locally available sender/receiver clocks. The authors also present the simple formal model of the algorithm. This formal model allows to make an elegant and short proof of correctness of the protocol.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Protein-protein interactions are fundamental to many biological processes. Experimental screens have identified tens of thousands of interactions and structural biology has provided detailed functional insight for select 3D protein complexes. An alternative rich source of information about protein interactions is the evolutionary sequence record. Building on earlier work, we show that analysis of correlated evolutionary sequence changes across proteins identifies residues that are close in space with sufficient accuracy to determine the three-dimensional structure of the protein complexes. We evaluate prediction performance in blinded tests on 76 complexes of known 3D structure, predict protein-protein contacts in 32 complexes of unknown structure, and demonstrate how evolutionary couplings can be used to distinguish between interacting and non-interacting protein pairs in a large complex. With the current growth of sequence databases, we expect that the method can be generalized to genome-wide elucidation of protein-protein interaction networks and used for interaction predictions at residue resolution.

In this short note, we show how to use concentration inequalities in order to build exact confidence intervals for the Hurst parameter associated with a one-dimensional fractional Brownian motion.