9Genetic data are often used to infer history, demographic changes or detect genes under se-10 lection. Inferential methods are commonly based on models making various strong assumptions: 11 demography and population structures are supposed a priori known, the evolution of the genetic 12 composition of a population does not affect demography nor population structure, and there is no 13 selection nor interaction between and within genetic strains. In this paper, we present a stochastic 14 birth-death model with competitive interaction to describe an asexual population, and we develop 15 an inferential procedure for ecological, demographic and genetic parameters. We first show how genetic diversity and genealogies are related to birth and death rates, and to how individuals com-17 pete within and between strains. This leads us to propose an original model of phylogenies, with 18 trait structure and interactions, that allows multiple merging. Second, we develop an Approxi-19 mate Bayesian Computation framework to use our model for analyzing genetic data. We apply 20 our procedure to simulated and real data. We show that the procedure give accurate estimate of 21 the parameters of the model. We finally carry an illustration on real data and analyze the genetic 22 diversity of microsatellites on Y-chromosomes sampled from Central Asia populations in order to 23 test whether different social organizations show significantly different fertility. 24

Genetic data are often used to infer demographic history and changes or detect genes under selection. Inferential methods are commonly based on models making various strong assumptions: demography and population structures are supposed a priori known, the evolution of the genetic composition of a population does not aect demography nor population structure, and there is no selection nor interaction between and within genetic strains. In this paper, we present a stochastic birth-death model with competitive interactions and asexual reproduction. We develop an inferential procedure for ecological, demographic and genetic parameters. We rst show how genetic diversity and genealogies are related to birth and death rates, and to how individuals compete within and between strains. This leads us to propose an original model of phylogenies, with trait structure and interactions, that allows multiple merging. Second, we develop an Approximate Bayesian Computation framework to use our model for analyzing genetic data. We apply our procedure to simulated data from a toy model, and to real data by analyzing the genetic diversity of microsatellites on Y-chromosomes sampled from Central Asia human populations in order to test whether dierent social organizations show signicantly dierent fertility.

We consider a C 3 family t → ft of C 4 Anosov diffeomorphisms on a compact Riemannian manifold M . Denoting by ρt the SRB measure of ft, we prove that the map t → θdρt is differentiable if θ is of the formfunction and a a regular value of g. We also require a transversality condition, namely that the intersection of the support of h with the level set {g(x) = a} is foliated by 'admissible stable leaves'. * matthieu.porte@yahoo.fr † This work was conducted in 2017 as a MSc thesis at UPMC, under the direction of V. Baladi (CNRS / IMJ-PRG), whom I thank for her guidance and her useful remarks at all stages of this work. I also thank P-A. Guihéneuf (UMPC / IMJ-PRG) for his helpful comments.

Abstract. This paper describes a methodology to produce a 7-classes land cover map of urban areas from very high resolution images and limited noisy labeled data. The objective is to make a segmentation map of a large area (a french department) with the following classes: asphalt, bare soil, building, grassland, mineral material (permeable artificialized areas), forest and water from 20cm aerial images and Digital Height Model.We created a training dataset on a few areas of interest aggregating databases, semi-automatic classification, and manual annotation to get a complete ground truth in each class.A comparative study of different encoder-decoder architectures (U-Net, U-Net with Resnet encoders, Deeplab v3+) is presented with different loss functions.The final product is a highly valuable land cover map computed from model predictions stitched together, binarized, and refined before vectorization.

Hyperspectral imaging, which consists in imaging a scene at a large number of wavelengths, has several applications, such as mineral identification, target detection, or gas concentration measurement. Most of the remote sensing missions would prefer to have compact instruments, and the ability to measure the information with a single acquisition (snapshot) may also be very interesting. Indeed, the information is not sensitive to the temporal variations of the scene: thus the acquisition of a three-dimensional (x,y,λ) hyperspectral data cubes of fast phenomenon (moving targets/gases) is possible. Furthermore, the hyperspectral image can be used in real time, and snapshot acquisition also reduces constraints on a scanning system. Among the concepts of "snapshot" hyperspectral imaging, the one proposed by Hirai (Hirai et al., Optical Review, 1, 205-207 (1994)) is very interesting. It relies on the association of a microlens array and a Fourier transform interferometer. The latter can be birefringent (Kudenov et al., Optics express, 20(16), 17973-17986 (2012)), which makes the system more compact and less sensitive to vibration by avoiding the use of a beam splitter (as with a Michelson interferometer for instance). Several designs of birefringent interferometers are possible, although the most compact solution is to use a birefringent interferometer with a fringe localization plane at a finite distance and accessible without a relay lens. With such an interferometer, for example a Nomarski prism, the image plane and the plane of localization of the fringes can be easily superposed on a detector. In this paper, the principle and the parameters that define the spectral/spatial performances of this snapshot hyperspectral imaging design are described. Several scenarios from visible to longwave infrared are presented to highlight the trade-off between spectral and spatial resolution. We then present a study of the propagation of spherical wavefronts through a birefringent interferometer by using 3D simulation and 2D analytical calculation. This tool allows us to quantitatively estimate the impact of the interferometer on the spatial quality of the image (aberration, transverse/axial shift of each channel) and evaluate the real interference pattern and the fringe visibility, i.e. the spectral quality, for the whole field-ofview.