Descripteur
Documents disponibles dans cette catégorie (18)
Ajouter le résultat dans votre panier
Visionner les documents numériques
Affiner la recherche Interroger des sources externes
Etendre la recherche sur niveau(x) vers le bas
Nonlinear unmixing of hyperspectral data using semi-nonnegative matrix factorization / Naoto Yokoya in IEEE Transactions on geoscience and remote sensing, vol 52 n° 2 (February 2014)
[article]
Titre : Nonlinear unmixing of hyperspectral data using semi-nonnegative matrix factorization Type de document : Article/Communication Auteurs : Naoto Yokoya, Auteur ; Jocelyn Chanussot, Auteur ; Akira Iwasaki, Auteur Année de publication : 2014 Article en page(s) : pp 1430 - 1437 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] analyse des mélanges spectraux
[Termes IGN] factorisation
[Termes IGN] image hyperspectrale
[Termes IGN] matrice
[Termes IGN] modèle non linéaireRésumé : (Auteur) Nonlinear spectral mixture models have recently received particular attention in hyperspectral image processing. In this paper, we present a novel optimization method of nonlinear unmixing based on a generalized bilinear model (GBM), which considers the second-order scattering of photons in a spectral mixture model. Semi-nonnegative matrix factorization (semi-NMF) is used for the optimization to process a whole image in matrix form. When endmember spectra are given, the optimization of abundance and interaction abundance fractions converge to a local optimum by alternating update rules with simple implementation. The proposed method is evaluated using synthetic datasets considering its robustness for the accuracy of endmember extraction and spectral complexity, and shows smaller errors in abundance fractions rather than conventional methods. GBM-based unmixing using semi-NMF is applied to the analysis of an airborne hyperspectral image taken over an agricultural field with many endmembers, and it visualizes the impact of a nonlinear interaction on abundance maps at reasonable computational cost. Numéro de notice : A2014-076 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2013.2245671 En ligne : https://doi.org/10.1109/TGRS.2013.2245671 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=32981
in IEEE Transactions on geoscience and remote sensing > vol 52 n° 2 (February 2014) . - pp 1430 - 1437[article]Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 065-2014021 RAB Revue Centre de documentation En réserve L003 Disponible
Titre : Bayesian and frequentist regression methods Type de document : Monographie Auteurs : Jon Wakefield, Auteur Editeur : Berlin, Heidelberg, Vienne, New York, ... : Springer Année de publication : 2013 Importance : 697 p. Format : 16 x 24 cm ISBN/ISSN/EAN : 978-1-4419-0925-1 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Statistiques
[Termes IGN] estimation bayesienne
[Termes IGN] fonction spline
[Termes IGN] inférence
[Termes IGN] méthode du maximum de vraisemblance (estimation)
[Termes IGN] méthode fondée sur le noyau
[Termes IGN] modèle de régression
[Termes IGN] modèle linéaire
[Termes IGN] modèle non linéaire
[Termes IGN] régression
[Termes IGN] théorème de BayesRésumé : (éditeur) Bayesian and Frequentist Regression Methods provides a modern account of both Bayesian and frequentist methods of regression analysis. Many texts cover one or the other of the approaches, but this is the most comprehensive combination of Bayesian and frequentist methods that exists in one place. The two philosophical approaches to regression methodology are featured here as complementary techniques, with theory and data analysis providing supplementary components of the discussion. In particular, methods are illustrated using a variety of data sets. The majority of the data sets are drawn from biostatistics but the techniques are generalizable to a wide range of other disciplines. While the philosophy behind each approach is discussed, the book is not ideological in nature and an emphasis is placed on practical application. It is shown that, in many situations, careful application of the respective approaches can lead to broadly similar conclusions. To use this text, the reader requires a basic understanding of calculus and linear algebra, and introductory courses in probability and statistical theory. The book is based on the author's experience teaching a graduate sequence in regression methods. The book website contains all of the code to reproduce all of the analyses and figures contained in the book. Note de contenu : 1- Inferential Approches
2- Independant data
3- Dependent data
4- Nonparametric modelingNuméro de notice : 25769 Affiliation des auteurs : non IGN Thématique : MATHEMATIQUE Nature : Monographie En ligne : https://doi.org/10.1007/978-1-4419-0925-1 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=94957 Diameter and death of whorl and interwhorl branches in Atlas cedar (Cedrus atlantica Manetti): a model accounting for acrotony / François Courbet in Annals of Forest Science, Vol 69 n° 2 (March 2012)
[article]
Titre : Diameter and death of whorl and interwhorl branches in Atlas cedar (Cedrus atlantica Manetti): a model accounting for acrotony Type de document : Article/Communication Auteurs : François Courbet, Auteur ; Jean-Christophe Hervé (1961-2017) , Auteur ; Etienne K. Klein, Auteur ; Francis Colin, Auteur Année de publication : 2012 Article en page(s) : pp 125 - 138 Note générale : bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Végétation
[Termes IGN] branche (arbre)
[Termes IGN] Cedrus atlantica
[Termes IGN] diamètre des arbres
[Termes IGN] houppier
[Termes IGN] modèle non linéaire
[Termes IGN] modèle numérique
[Termes IGN] nutrition végétaleRésumé : (auteur) Branch size and branch status (dead or alive) are important characteristics closely related to tree growth and wood quality. The aim of this study was to design models for the diameter and status of branches in Atlas cedar (Cedrus atlantica Manetti). The models were developed from data collected on a set of 32 trees with a wide range of heights (from 3 to 36 m), girths (from 13 to 226 cm), and ages (from 20 to 95 years). A single general segmented model was designed for both whorl and interwhorl branch diameter, taking into account the tree and annual growth unit random effects. The model’s “potential x reducer” form describes the maximum branch diameter profile along the tree and the acrotonic gradient observed in annual shoots. The diameter and status of every branch were modeled based on the vertical position on the trunk and on the height of the base of the living crown. The tree diameter and the branch diameter were used as additional explanatory variables in the branch diameter model and the branch status model, respectively. The model structure is sufficiently general to be suited after re-parameterization for many coniferous species with interwhorl branches such as Spruces, Firs, and Larches. Numéro de notice : A2012-718 Affiliation des auteurs : IFN+Ext (1958-2011) Thématique : FORET Nature : Article DOI : 10.1007/s13595-011-0156-1 Date de publication en ligne : 06/12/2011 En ligne : http://dx.doi.org/10.1007/s13595-011-0156-1 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83725
in Annals of Forest Science > Vol 69 n° 2 (March 2012) . - pp 125 - 138[article]Modelos armonicos no lineales para series temporales geodéticas = Non-linear harmonic models for geodetic time series / P.A. Martinez-Ortiz (2011)
Titre : Modelos armonicos no lineales para series temporales geodéticas = Non-linear harmonic models for geodetic time series Type de document : Thèse/HDR Auteurs : P.A. Martinez-Ortiz, Auteur ; J.M. Fernadiz Leal, Directeur de thèse Editeur : Alicante : Escuella politécnica superior Année de publication : 2011 Importance : 402 p. Format : 21 x 30 cm Note générale : Bibliographie Langues : Espagnol (spa) Descripteur : [Vedettes matières IGN] Géodésie
[Termes IGN] analyse harmonique
[Termes IGN] analyse spectrale
[Termes IGN] bruit blanc
[Termes IGN] bruit rose
[Termes IGN] géocentre
[Termes IGN] marée terrestre
[Termes IGN] masse de la Terre
[Termes IGN] Matlab
[Termes IGN] modèle non linéaire
[Termes IGN] positionnement par GPS
[Termes IGN] série temporelle
[Termes IGN] système de référence géodésiqueIndex. décimale : THESE Thèses et HDR Résumé : (Auteur) The dissertation addresses the development of new methods and software for the spectral analysis of scalar or vectorial time series, with emphasis in the applications of geodetic interest. The starting point can be placed in the method introduced by Harada and Fukushima for the non-linear analysis of time series, which allows the recursive detection of frequencies and their associated amplitudes and phases as well as the secular mixed Fourier terms when found in the signal. That method is extended in different ways, allowing the treatment of series affected by auto correlated noise with a power law, either evenly or unevenly spaced. This is made both at the level of frequency detection and non-linear fitting. Reduction of the computational overhead is also obtained. The theoretical work is accompanied by the developing of comprehensive and specialized software for such non-linear harmonic analysis of time series using the MATLAB programming language. Much of the tools we can find today for analyzing these periodic time series are valid only for certain types whereas the programs in the thesis can be applied to oddly spaced series in the presence of combinations of white and flicker noise. The new methods and routines are used for analyzing some interesting series as those ones describing the celestial pole offsets, the geocentre variations due to the redistribution of water mass on the Earth's surface, the excess of the length of day, continental water flux and the positions of GPS stations, among others. We estimate harmonic models that explain each one of these phenomena in the considered time domain and allow us to draw conclusions of their behavior. Note de contenu : Agradecimientos
Resumen
Abstract
I MODELOS ARMÓNICOS NO LINEALES PARA SE RÍES TEMPORALES GEODÉTICAS
1. Introducción
1.1. Objetivos y metodología
1.2. Contenidos
2. El problema de la detección de señales
2.1. Definición de serie temporal
2.1.1. Componentes de una serie temporal
2.1.2. Componente de ruido
2.2. Técnicas para el estudio de series temporales
2.2.1. Análisis clásico
2.2.2. Análisis espectral
2.2.3. Análisis wavelet
2.3. El problema de la detección de señales
2.3.1. Periodograma de Lomb
2.3.2. Dominio de frecuencia
3. Análisis armónico no lineal
3.1. Introducción
3.2. Descripción del método
3.2.1. Función objetivo y funciones base
3.2.2. Solución mínimo cuadrática
3.2.3. Optimización no lineal: Algoritmo BFGS
3.2.4. Extracción de frecuencias
3.3. Incertidumbre
3.4. Generalización
3.5. Tratamiento de los términos periódicos de corta frecuencia
4. Variaciones del polo celeste
4.1. Modelo de precesión-nutación IAU1980
4.1.1. Introducción
4.1.2. Descripción de los datos
4.1.3. Características del análisis y resultados
4.1.4. Conclusiones
4.2. Modelo de precesión-nutación IAU2000
4.2.1. Introducción
4.2.2. Descripción de los datos
4.2.3. Características del análisis y resultados
4.2.4. Conclusiones
4.3. Modelos dinámicos para la predicción a corto plazo de (óV óe)
5. Variaciones geocéntricas causadas por el flujo de agua continental 101
5.1. Introducción
5.2. Descripción de los datos
5.3. Características del análisis y resultados
5.4. Conclusiones
6. Modelos espacio-temporales para el flujo de agua continental
6.1. Introducción
6.2. Descripción de datos
6.3. Características del análisis
6.4. Resultados
6.5. Conclusiones
7. Estudio armónico del exceso en la duración del día
7.1. Introducción
7.2. Descripción de los datos
7.3. Características del análisis y resultados
7.4. Conclusiones
8. Ruido
8.1. Tipología básica
8.2. Matrices de covarianza
8.2.1. Matriz de covarianzas para un ruido blanco
8.2.2. Matriz de covarianzas para un ruido parpadeante
8.2.3. Matriz de covarianzas para un paseo aleatorio
8.3. Relación ruido-periodograma
8.4. Ruido en las observaciones GPS
9. Algoritmo FHAST
9.1. Introducción
9.2. Función objetivo
9.3. Modelo estocástico
9.3.1. Estimación de un índice espectral
9.3.2. Componente residual como combinación de varios ruidos. Es-timación de la frecuencia de transición,
9.4. Modelo funcional
9.5. Estimación de la componente de varianza
9.5.1. Condición de no negatividad
9.6. Extensión del periodograma
9.6.1. Aceleración del cálculo del periodograma
9.7. Incertidumbre
9.7.1. Parámetros lineales y no lineales del modelo funcional
9.7.2. Parámetros del modelo estocástico
9.8. Criterios de parada algorítmica
9.9. Entramado algorítmico
9.10. Simulación
10.Estudio de las series de posiciones de estaciones GPS
10.1. Introducción
10.2. Análisis de las series temporales residuales
10.3. Resultados y discusión
10.4. Conclusiones
11.Conclusiones y perspectivas
11.1. Conclusiones
11.2. Perspectivas
II EXTENDED SUMMARY: NON-LINEAR HARMO NIC MODELS FOR GEODETIC TIME SERIES
S.1. Introduction
S.2. The signal detection problem
S.2.1. Definition of time series
S.2.2. Spectral analysis
S.2.3. The signal detection problem
S.3. Non-linear harmonic analysis
S.4. Celestial Pole Offsets
S.4.1. IAU1980 Pole Offsets
S.4.2. IAU2000 Pole Offsets
S.4.3. Dynamic models for (óV,óe) prediction
S.5. Geocenter variations caused by continental water flux
S.5.1. Introduction
S.5.2. Data
S.5.3. Analysis and results
S.5.4. Conclusions
S.6. Spatio-temporal models for continental water flux
S.6.1. Introduction
S.6.2. Data
S.6.3. Methods
S.6.4. Results
S.6.5. Conclusions
S.7. Harmonio study of the length of day
S.7.1. Introduction
S.7.2. Data
S.7.3. Analysis and results
S.7.4. Conclusions
S.8. Noise
S.8.1. Typology
S.8.2. Covariance matrices
S.8.3. Relationship noise-periodogram
S.8.4. Noise in GPS observations
S.9. FHAST algorithm
S.9.1. Introduction
S.9.2. Stochastic model
S.9.3. Functional model
S.9.4. Component variance estimation
S.9.5. Modification of the periodogram
S.9.6. Stop criteria
S9.7. Algorithm
S.10. Study of the position time series of GPS stations
S.10.1.Introduction
S.10.2. Analysis and results
S.10.S.Conclusions
S.11 Conclusions and outlook
S.11.1.Conclusions
S.11.2. Outlook
III APÉNDICES
A. Ley de propagación del error
B, Acrónimos, abreviaturas y unidades
C. Modelos armónicos no lineales de algunas estaciones GPS
BibliografíaNuméro de notice : 10519 Affiliation des auteurs : non IGN Thématique : POSITIONNEMENT Nature : Thèse française Note de thèse : Bibliographie nature-HAL : Thèse DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=45141 Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 10519-01 30.60 Livre Centre de documentation Géodésie Disponible Automated derivation of bathymetric information from multi-spectral satellite imagery using a non-linear inversion model / H. Su in Marine geodesy, vol 31 n° 4 (December 2008)
[article]
Titre : Automated derivation of bathymetric information from multi-spectral satellite imagery using a non-linear inversion model Type de document : Article/Communication Auteurs : H. Su, Auteur ; H. Liu, Auteur ; W.D. Heymann, Auteur Année de publication : 2008 Article en page(s) : pp 281 - 298 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique
[Termes IGN] données bathymétriques
[Termes IGN] étalonnage de modèle
[Termes IGN] fond marin
[Termes IGN] image à résolution métrique
[Termes IGN] image Ikonos
[Termes IGN] image multibande
[Termes IGN] modèle d'inversion
[Termes IGN] modèle non linéaire
[Termes IGN] optimisation (mathématiques)Résumé : (Auteur) Most previous studies utilized a log-linear regression model to invert multi-spectral images into bathymetric data. Based on the Levenberg-Marquardt optimization algorithm, we developed an automated method for calibrating the parameters for a non-linear inversion model. Our method has been successfully applied to an IKONOS multispectral image. We compared depth data derived from our model to those estimated using a conventional log-linear inversion model. Bathymetric data derived from the non-linear inversion model are slightly more accurate and stable, particularly for deeper benthic habitats, than those derived from a conventional log-linear model although their overall performances are very similar. Copyright Taylor & Francis Numéro de notice : A2008-515 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.1080/01490410802466652 En ligne : https://doi.org/10.1080/01490410802466652 Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=29584
in Marine geodesy > vol 31 n° 4 (December 2008) . - pp 281 - 298[article]Réservation
Réserver ce documentExemplaires(1)
Code-barres Cote Support Localisation Section Disponibilité 230-08041 RAB Revue Centre de documentation En réserve L003 Disponible A quaternion-based geodetic datum transformation algorithm / Y.Z. Shen in Journal of geodesy, vol 80 n° 5 (August 2006)PermalinkAusgleichungsrechnung in nichtlinearen Modellen / P. Lohse (1994)Permalink