33 research outputs found
Annotated translation: Concevoir un jeu vidéo (Marc Albinet: Concevoir un jeu vidéo. Les méthodes et les outils des professionnels expliqués à tous. FYP éditions, Roubaix, 2015.)
No full textMy bachelor thesis consists of two parts. The first part consists of a translation of 20 standard pages of the book Concevoir un jeu vidéo. Les méthodes et les outils des professionnels expliqués à tous. by the French author and video game developer Marc Albinet. The second part consist of an analysis of the original, a description of the chosen translation strategy, the problems and their solutions, and the translation shifts I resorted to
Vertical Structure of P-Band Temporal Decorrelation at the Paracou Forest: Results From TropiScat
No full textAnnotated translation: Concevoir un jeu vidéo (Marc Albinet: Concevoir un jeu vidéo. Les méthodes et les outils des professionnels expliqués à tous. FYP éditions, Roubaix, 2015.)
No full textMy bachelor thesis consists of two parts. The first part consists of a translation of 20 standard pages of the book Concevoir un jeu vidéo. Les méthodes et les outils des professionnels expliqués à tous. by the French author and video game developer Marc Albinet. The second part consist of an analysis of the original, a description of the chosen translation strategy, the problems and their solutions, and the translation shifts I resorted to.Má bakalářská práce je tvořena dvěma části. První částí sestává z překladu 20 normostran díla Concevoir un jeu vidéo. Les méthodes et les outils des professionnels expliqués à tous. francouzského autora a videoherního vývojáře Marca Albineta. Druhá část obsahuje analýzu originálu, popis zvolené překladatelské strategie, problémů a jejich řešení a překladatelských posunů, ke kterým jsem se uchýlil.Ústav translatologieInstitute of Translation StudiesFaculty of ArtsFilozofická fakult
FoPen Man-Made Objects Detection with P-band SAR Sub-Aperture Analysis
No full textInternational audienceFoPen (Foliage Penetration) SAR systems, especially those operating at low frequencies, are effective for detecting man-made objects in dense forests. This study leverages a single SAR image to distinguish between the strong volume scattering from foliage and the signals of concealed man-made objects, which are typically non-isotropic along the view angle, unlike the azimuthally invariant forest or vegetation. We propose a twostep strategy: firstly, a sliding sublook analysis maintains the azimuthal bandwidth maximally wide for each sublook, producing SAR images with reduced azimuthal resolution but varied central along-track components of the wave-vector, associated with the target's 'aspect angle'. Secondly, a homogeneity test-specifically, the coefficient of variation-analyzes the aspect angle dimension of the SAR stack. Preliminary results using P-band datasets from the TropiSAR campaign in Paracou demonstrate the effectiveness of our approach in detecting concealed vehicles and infrastructures.La détection d'objets fabriqués par l'homme dans des forêts denses représente un défi pour les applications de sécurité et de surveillance. Les systèmes SAR FoPen (Penetration de Feuillage) opérant à basse fréquence représentent une très bonne solution pour une telle tâche. Le principal défi est de discriminer avec précision entre 1) la forte diffusion volumique du feuillage associée aux interactions sol-tronc qui rétrodiffusent vers le radar et 2) le signal des objets fabriqués par l'homme cachés sous le couvert forestier. Dans les forêts très denses, comme les forêts tropicales, il est généralement nécessaire de recourir à des algorithmes de détection de changement pour détecter un objet qui est arrivé ou qui a quitté la scène acquise. Dans cette contribution, nous considérons qu'une seule image SAR de la scène est disponible. Nous supposons que les objets fabriqués par l'homme présentent une forte variabilité des sections efficaces radar en fonction de l'angle de vue et, d'autre part, que les arbres ou les paquets d'arbres conservent un rétrodiffusé relativement homogène en fonction de l'angle de vue. Notre stratégie consiste en deux étapes principales : la première étape effectue une analyse sublook glissante en gardant la bande passante azimutale aussi large que possible pour chaque sublook, conduisant à un ensemble d'images SAR avec une résolution azimutale réduite et une composante centrale variable le long de la piste du vecteur d'onde que nous relions à la dimension 'angle d'aspect' de la cible. La seconde étape consiste en un test d'homogénéité, tel que le coefficient de variation le long de la dimension de l'angle d'aspect de l'empilement SAR susmentionné généré. Nous montrons quelques résultats préliminaires sur la détection de véhicules et d'infrastructures dissimulés en utilisant des ensembles de données en bande P acquises à Paracou lors de la campagne TropiSAR
Investigations on the Effect of HD Processing in Land Cover Classification
No full textISSN:1682-1750ISSN:2194-9034ISSN:1682-1777ISSN:2194-9034ISSN:1682-177
Ecologie linguistique des milieux littoraux: étude de cas dialectométrique sur les langues et variétés dialectales du Golfe du Lion, selon les données de Louis Michel (1964)
No full textInternational audienceLes dades geolingüístiques o dialectològiques sovint són riques en indicis sobre els esquemes d'interacció en geografia humana i en la història d'una regió o d'un entorn natural habitat per una població rural establerta de més o menys temps. Es presentarà aquí una costa ben lluny de les costes de la Manega, però que podria inspirar treballs analògics: el Golf del Lleó, en la seva dimensió lingüística, tal com es recull en un document molt valuós: l'atles costaner de Louis Michel (1964). L'autor hi presenta costes i grups humans vinculats per una professió (la pesca), repartits en una àmplia franja de graus, estanys i lidos als peus dels monts del Llenguadoc i de Provença. Les dades de camp de Louis Michel, recollides entre 1954 i 1964 entre els pescadors del Golf del Lleó, són insuperables: tenen una gran solidesa empírica i s'han recollit en les llengües del corpus (català i occità/provençal). S'ha realitzat una captura numèrica de dades dialectals d'aquest estat documental sobre les varietats del català rossellonès i de l'occità a la Universitat de Montpeller 3, permetent oferir una "fotografia" en un moment donat, documentada in extremis abans de l'assimilació sociolingüística dels anys 1980-2010. Tenen un alt contingut en indexicalitat sociohistòrica sobre les dinàmiques lingüístiques observables, especialment mitjançant un tractament dialectomètric (distància d'edició: programari Gabmap 2011-: vegeu Nerbonne i altres 2011; Nerbonne i Heeringa, 2001, Nerbonne i altres 2005; Nerbonne i Kretzschmar 2003; consulteu Séguy 1971, 1973 per a les bases de la dialectologia quantitativa, i Polian i Léonard per a l'aplicació a un àmbit no romànic). S'incorporarà la zona mediterrània estudiada en el seu context occità (Léonard, Brun-Trigaud i Picard 2023, Léonard i Brun-Trigaud 2022), i es revisitaran les nocions de "modalitats" i de "transició" de Ricardo Carvalho Calero (vegeu Dubert-Garcia 2021) a la llum d'aquest estudi de cas. Es podria plantejar un enfocament similar sobre les dinàmiques lingüístiques a Normandia i a la costa de la Manega. Això contrastaria i faria interactuar parlars angevins, normands, picards, ille-de-francesos i flamencs (Brasseur 2016, Loriot 1967). La zona és molt menys lacunar i sobretot, més industrialitzada. Cal construir una metodologia en ecologia diasistèmica, al voltant d'espais tan complexos però relativament ben delimitats com el Golf del Lleó.Geolinguistic or dialectological data often provide rich insights into interaction patterns in human geography and the history of a region or natural environment inhabited by a rural population for varying lengths of time. This paper will present a coastline quite distant from the shores of the English Channel but which could inspire similar research: the Gulf of Lion, in its linguistic dimension, as captured in a very valuable document: Louis Michel's coastal atlas (1964). The author presents coastlines and human groups connected by a profession (fishing), spread out along a vast stretch of lagoons, ponds, and barrier islands at the foot of the Languedoc and Provence mountains. Louis Michel's field data, collected from Gulf of Lion fishermen between 1954 and 1964, are unparalleled: they exhibit great empirical solidity and were collected in the languages of the corpus (Catalan and Occitan/Provençal). A digital capture of dialectal data from this documentary record on the varieties of Roussillon Catalan and Occitan has been carried out at the University of Montpellier 3, allowing for a "snapshot" at a given moment, documented just in time before the sociolinguistic assimilation of the 1980s-2010s. They have a high level of sociohistorical indexicality regarding observable linguistic dynamics, especially through dialectometric analysis (editing distance: Gabmap software 2011-; see Nerbonne et al. 2011; Nerbonne and Heeringa, 2001; Nerbonne et al. 2005; Nerbonne and Kretzschmar 2003; see Séguy 1971, 1973 for the foundations of quantitative dialectology, and Polian and Léonard for its application to a non-Romance domain). The Mediterranean area studied will be situated in its Occitan context (Léonard, Brun-Trigaud, and Picard 2023, Léonard and Brun-Trigaud 2022), and the notions of "modalities" and "transition" by Ricardo Carvalho Calero will be revisited (see Dubert-Garcia 2021) in light of this case study. A similar approach to linguistic dynamics in Normandy and along the English Channel coast could be considered. This would involve contrasting and observing the interaction of Angevin, Norman, Picard, Île-de-France, and Flemish dialects (Brasseur 2016, Loriot 1967). The area is much less lagoonal and, above all, more industrialized. A methodology in diasystemic ecology remains to be constructed, centered around spaces that are equally complex but relatively well-defined, such as the Gulf of Lion.Les données géolinguistiques ou dialectologiques sont souvent riches en indices sur les schèmes d’interaction en géographie humaine et dans l’histoire d’une région ou d’un milieu naturel habité par une population rurale implantée de plus ou moins longue date. On présentera ici un littoral bien éloigné des rives de la Manche, mais qui pourrait y inspirer des travaux analogues : le Golfe du Lion, dans sa dimension linguistique, telle que saisie dans un document très précieux : l’atlas côtier de Louis Michel (1964). L’auteur y présente des rives et des groupes humains liés par un métier (la pêche), répartis sur un vaste cordon de graus, d’étangs, de lidos au pied des monts du Languedoc et de Provence. Les données de terrain de Louis Michel, recueillies en 1954-64 auprès des pêcheurs du Golfe du Lion, sont insurpassables : elles sont d’une grande solidité empirique, et ont été recueillies dans les langues du corpus (catalan et occitan/provençal). Une saisie numérique de données dialectales de cet état documentaire sur les variétés de catalan roussillonnais et d’occitan a été réalisée à l’Université de Montpellier 3, permettant de proposer une « photographie » à un moment donné, documentée in extremis avant l’assimilation sociolinguistique des années 1980-2010. Elles revêtent une haute teneur en indexicalité sociohistorique sur les dynamiques langagières observables, notamment par un traitement dialectométrie (distance d’édition : logiciel Gabmap 2011- : v. Nerbonne & al. 2011 ; Nerbonne & Heeringa, 2001, Nerbonne & al. 2005 ; Nerbonne & Kretzschmar 2003 ; cf. Séguy 1971, 1973 pour les fondations de la dialectologie quantitative, et Polian & Léonard pour l’application à un domaine non roman). On insèrera la zone méditerranéenne étudiée dans son contexte occitan (Léonard, Brun-Trigaud & Picard 2023, Léonard & Brun-Trigaud 2022), et on revisitera les notions de « modalités » et de « transition » de Ricardo Carvalho Calero (v. Dubert-Garcia 2021) à la lumière de cette étude de cas. Une approche analogue sur les dynamiques langagières en Normandie et sur le littoral de la Manche pourrait être envisagée. Elle ferait contraster et verrait interagir parlers angevins, normands, picards, île-de-franciens et flamands (Brasseur 2016, Loriot 1967). La zone est bien moins lagunaire et surtout, plus industrialisée. Une méthodologie en écologie diasystémique reste à construire, autour d’espaces aussi complexes mais relativement bien délimités, comme le Golfe du Lion.I dati geolinguistici o dialettologici spesso forniscono ricche indicazioni sui modelli di interazione nella geografia umana e nella storia di una regione o di un ambiente naturale abitato da una popolazione rurale insediata da più o meno tempo. Qui si presenta una costa ben lontana dalle rive della Manica, ma che potrebbe ispirare lavori analoghi: il Golfo del Leone, nella sua dimensione linguistica, come catturato in un documento molto prezioso: l'atlante costiero di Louis Michel (1964). L'autore presenta coste e gruppi umani legati da un mestiere (la pesca), distribuiti lungo una vasta fascia di graus, laghi, lidi ai piedi dei monti del Linguadoca e della Provenza. I dati sul campo raccolti da Louis Michel dai pescatori del Golfo del Leone tra il 1954 e il 1964 sono insuperabili: sono di grande solidità empirica e sono stati raccolti nelle lingue del corpus (catalano e occitano/provenzale). È stata effettuata una digitalizzazione dei dati dialettali di questo documento sulle varietà del catalano roussillonese e dell'occitano presso l'Università di Montpellier 3, consentendo di offrire una "fotografia" in un determinato momento, documentata all'ultimo minuto prima dell'assimilazione sociolinguistica degli anni '80 e '10. Questi dati presentano un alto contenuto di indicizzazione socio-storica sulle dinamiche linguistiche osservabili, in particolare attraverso l'analisi dialettometrica (distanza di Levenshtein: software Gabmap 2011-; vedi Nerbonne et al. 2011; Nerbonne e Heeringa, 2001; Nerbonne et al. 2005; Nerbonne e Kretzschmar 2003; cf. Séguy 1971, 1973 per le fondamenta della dialettologia quantitativa, e Polian e Léonard per l'applicazione a un campo non romanico). La zona mediterranea studiata verrà inserita nel contesto occitano (Léonard, Brun-Trigaud e Picard 2023, Léonard e Brun-Trigaud 2022), e saranno rivalutati i concetti di "modalità" e "transizione" di Ricardo Carvalho Calero (vedi Dubert-Garcia 2021) alla luce di questo studio di caso. Un approccio simile alle dinamiche linguistiche in Normandia e lungo la costa della Manica potrebbe essere preso in considerazione. Questo comporterebbe il contrasto e l'interazione tra i dialetti angevini, normanni, piccardi, dell'Île-de-France e fiamminghi (Brasseur 2016, Loriot 1967). La zona è molto meno lagunare e soprattutto più industrializzata. Rimane da costruire una metodologia in ecologia diasistemica, attorno a spazi altrettanto complessi ma relativamente ben definiti, come il Golfo del Leone.Las donadas géolinguistiques o dialectologicas son sovent ricas en indicis sus las schèmes d'interaccion en geografia umana e dins l'istòria d'una region o d'un mièg natural abitat per una populacion rurala implantada en mai o mens longa data. Se presentarà aicí un litoral plan alunhat de las ribas de la Marga, mas que i poiriá inspirar de las òbras analògas : lo gòlfe del Leon, dins sa dimension lingüistica, tala coma sasida dins un document plan preciós : l'atlàs costièr de Louis Michel (1964). L'autor i presenta de ribas e dels grops umans ligats per un mestièr (la pesca), despartits sus un vast cordon de graus, d'estanhs, de lidos al pè dels monts de Lengadòc e de Provença. Las donadas de terren de Louis Michel, reculhidas en 1954-64 al près dels pescaires del gòlfe del Leon, son insuperables : son d'una granda soliditat empirica, e son estadas reculhidas dins las lengas del còrpus (catalan e occitan/provençal). Una sasida numerica de donadas dialectalas d'aquel estat documentari sus las varietats de catalan rosselhonés e d'occitan es estada realizada a l'Universitat de Montpelhièr 3, permetent de prepausar una « fotografia » a un moment balhat, documentada in extremis abans l'assimilacion sociolingüistica de las annadas 1980-2010. Revestisson una nauta tenor en indexicalité sociohistorique sus las dinamicas lengatgièras observablas, sustot per un tractament dialectométrie (distància d'edicion : logicial Gabmap 2011- : v. Nerbonne & al. 2011 ; Nerbonne & Heeringa, 2001, Nerbonne & al. 2005 ; Nerbonne & Kretzschmar 2003 ; cf. Séguy 1971, 1973 per las fondacions de la dialectologia quantitativa, e Polian & Léonard per l'aplicacion a un domeni non romanic). S'inserirà la zòna mediterranèa estudiada dins son contèxt occitan (Léonard, Brun-Trigaud & Picard 2023, Léonard & Brun-Trigaud 2022), e se tornar las nocions de « modalitats » e de « transicion » de Ricardo Carvalho Calero (v. Dubert-Garcia 2021) al lum d'aquel estudi de cas. Un apròchi analòg sus las dinamicas lengatgièras en Normandia e sul litoral de la Marga poiriá èsser considerada. Fariá contrastar e veiriá interagir parlars angevins, normands, picards, illa-de-francians e flamencs (Brasseur 2016, Auriòl 1967). La zòna es fòrça mens lagunària e sustot, mai industrializada. Una metodologia en ecologia diasystémique rèsta a bastir, a l'entorn d'espacis tanben complèxes mas relativament plan delimitats, coma lo gòlfe del Leon
Use of near-infrared spectroscopy in the investigation of brain activation during cognitive aging: A systematic review of an emerging area of research
No full textInternational audienceThe cognitive neuroscience of aging is a growing and stimulating research area. The development of neuroimaging techniques in the past two decades has considerably increased our understanding of the brain mechanisms that might underlie cognitive performance and resulting changes due to normal aging. Beside traditional metabolic neuroimaging techniques, such as Positron Emission Tomography and functional Magnetic Resonance Imaging, near infrared spectroscopy (NIRS), an optical imaging technique allowing to monitor real-time cerebral blood oxygenation, has gained recent interest in this field. The aim of the present review paper, after briefly presenting the NIRS technique, is to review and to summarize the recent results of neuroimaging studies using this technique in the field of cognitive aging. The reviewed literature shows that, despite low spatial resolution and cerebral depth penetration, this technique provides consistent findings on the reduced hemodynamic activity as a function of chronological age, mainly in the prefrontal cortex. Important moderators of brain hemodynamics, such as cognitive load, subjects’ characteristics and experimental conditions, for which the NIRS technique is sensitive, are discussed. Strengths and weaknesses of functional NIRS in the field of cognitive aging are presented and finally, novel perspectives of research are proposed
Machine-Learning Applications for the Retrieval of Forest Biomass from Airborne P-Band SAR Data
No full textThis study aimed at evaluating the potential of machine learning (ML) for estimating forest biomass from polarimetric Synthetic Aperture Radar (SAR) data. Retrieval algorithms based on two different machine-learning methods, namely Artificial Neural Networks (ANNs) and Supported Vector Regressions (SVRs), were implemented and validated using the airborne polarimetric SAR data derived from the AfriSAR, BioSAR, and TropiSAR campaigns. These datasets, composed of polarimetric airborne SAR data at P-band and corresponding biomass values from in situ and LiDAR measurements, were made available by the European Space Agency (ESA) in the framework of the Biomass Retrieval Algorithm Inter-Comparison Exercise (BRIX). The sensitivity of the SAR measurements at all polarizations to the target biomass was evaluated on the entire set of data from all the campaigns, and separately on the dataset of each campaign. Based on the results of the sensitivity analysis, the retrieval was attempted by implementing general algorithms, using the entire dataset, and specific algorithms, using data of each campaign. Algorithm inputs are the SAR data and the corresponding local incidence angles, and output is the estimated biomass. To allow the comparison, both ANN and SVR were trained using the same subset of data, composed of 50% of the available dataset, and validated on the remaining part of the dataset. The validation of the algorithms demonstrated that both machine-learning methods were able to estimate the forest biomass with comparable accuracies. In detail, the validation of the general ANN algorithm resulted in a correlation coefficient R = 0.88, RMSE = 60 t/ha, and negligible BIAS, while the specific ANN for data obtained R from 0.78 to 0.94 and RMSE between 15 and 50 t/ha, depending on the dataset. Similarly, the general SVR was able to estimate the target parameter with R = 0.84, RMSE = 69 t/ha, and BIAS negligible, while the specific algorithms obtained 0.22 ≤ R ≤ 0.92 and 19 ≤ RMSE ≤ 70 (t/ha). The study also pointed out that the computational cost is similar for both methods. In this respect, the training is the only time-demanding part, while applying the trained algorithm to the validation set or to any other dataset occurs in near real time. As a final step of the study, the ANN and SVR algorithms were applied to the available SAR images for obtaining biomass maps from the available SAR images
The ESA Sentinel-1 Normalized Radar Backscatter Product
No full textThe current family of Copernicus Sentinel-1 (European Space Agency / European Commission) products primarily contains Level-1 Single Look Complex (SLC) and Ground Range Detected (GRD) products [1], which inherit their definition from the European heritage Synthetic Aperture Radar (SAR) satellite missions ERS-1, -2 and ENVISAT. These products have over the years proven to be reliable high-quality data sources since the start of Sentinel-1A in 2014 and users largely benefit from the open and free data policy of the Copernicus program. This has led to Sentinel-1 products being routinely used in several operational applications as well as enlarging the user base of SAR data in general.
However, the rapid increase of data volume is presenting a challenge to many users who still want to exploit this wealth of information but lack the resources for the processing needed to convert these Level-1 products to interoperable geoinformation. Cloud exploitation of data offers opportunities for accelerated gain of knowledge but requires new strategies of data management and provision.
In this context, the term Analysis Ready Data (ARD) has been coined and several activities have indicated the potential of enlarging the Copernicus product family by such ARD products. However, no agreement has yet been reached on the definition of ARD as different user communities perform very different analyses and thus have different understandings of analysis readiness.
With the aim to standardize different categories of ARD, the Committee on Earth Observation Satellites (CEOS) has set up the CEOS Analysis Ready Data for Land initiative (CARD4L) [2], which has recently been renamed to simply CEOS ARD (CARD) to accommodate ocean products. Several product specifications have been defined to provide guidelines on how to best process and organize data to serve as many use cases as possible with the respective products. One such product is Normalized Radar Backscatter (NRB) [3].
However, the CARD NRB specification can also be thought of as a guideline rather than a specification. It gives recommendations on what information should be included in a dataset but gives no concrete specifications on how this information should be created. For this reason, data providers are required to translate these guidelines into actual products with the software of choice to then reach out to CEOS for assessment of CARD compliance.
Recently, ESA has defined its own prototype Sentinel-1 NRB product [4]. The aim is to be fully aligned with the CARD certification, offering a high-quality radiometrically terrain corrected SAR backscatter product with all relevant general metadata and per-pixel metadata (ancillary data layers). It is intended as a global and consistently processed product achieving the highest possible quality. It is designed to be complete to satisfy requirements of users with a prioritized focus on the Copernicus Services. Ultimately, it intends to significantly lower the effort for processing for the SAR user community. Core characteristics are the structuring into the Military Grid Reference System (MGRS) tile grid for interoperability with Sentinel-2 ARD products, storage in Cloud Optimized GeoTIFFs and the provision of Spatiotemporal Asset Catalogue (STAC) metadata.
We present the characteristics of this S1-NRB product, give insight into current work towards extending the product to Ocean Radar Backscatter (ORB), and outline the DLR plans of globally generating the product on the terrabyte High Performance Data Analytics (HPDA) platform. These activities are expected to contribute to the further definition of a Copernicus Sentinel-1 NRB product.
This type of product is now baseline in the frame of the Copernicus Expansion and Sentinel Next Generation and is hence planned for further future missions
