Centre de documentation
scientifique

Multilayer NMF for blind unmixing of hyperspectral imagery with additional constraints / L. Chen in Photogrammetric Engineering & Remote Sensing, PERS, vol 83 n° 4 (April 2017)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 83 n° 4 (April 2017) . - pp 307 - 316 Titre : Multilayer NMF for blind unmixing of hyperspectral imagery with additional constraints Type de document : Article/Communication Auteurs : L. Chen, Auteur ; Shengbo Chen, Auteur ; Xulin Guo, Auteur Année de publication : 2017 Article en page(s) : pp 307 - 316 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse des mélanges spectraux [Termes IGN] calcul matriciel [Termes IGN] contrainte spectrale [Termes IGN] factorisation de matrice non-négative [Termes IGN] filtrage du bruit [Termes IGN] image hyperspectrale [Termes IGN] processus de hiérarchisation analytique [Termes IGN] programmation par contraintes [Termes IGN] réflectance Résumé : (Auteur) Due to the coincidence of hyperspectral reflectance nonnegativity (and its corresponding abundance) with nonnegative matrix factorization (NMF) methods, NMF has been widely applied to unmix hyperspectral images in recent years. However, many local minima persist because of the nonconvexity of the objective function. Thus, the nonnegativity constraint is not sufficient and additional auxiliary constraints should be applied to objective functions. In this paper, a new approach we call constrained multilayer NMF (CMLNMF), is proposed for hyperspectral data. In this approach, the mixed spectra are regarded as endmember signatures that has been contaminated by multiplicative noise. The purpose of CMLNMF is to eliminate noise by hierarchical processing until the endmember spectra are obtained. Also, the hierarchical processing is self-adaptive to make the algorithm more effective. Furthermore, in each layer two constraints are implemented on the objective function. One is sparseness on the abundance matrix and the other is minimum volume on the spectral matrix. The hierarchical processing separates the abundance matrix into a series of matrices that make the characteristic of sparseness more obvious and meaningful. The proposed algorithm is applied to synthetic data and real hyperspectral data for quantitative evaluation. According to the comparison with other algorithms, CMLNMF has better performance and provides effective solutions for blind unmixing of hyperspectral image data. Numéro de notice : A2017-112 Affiliation des auteurs : non IGN Thématique : IMAGERIE/INFORMATIQUE/MATHEMATIQUE Nature : Article DOI : 10.14358/PERS.83.4.307 En ligne : https://doi.org/10.14358/PERS.83.4.307 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=84590 [article]

WebGIS performance tests / H. Dai in GIM international, vol 26 n° 1 (January 2012)

Public [article]  inGIM international > vol 26 n° 1 (January 2012) . - pp 23 - 27 Titre : WebGIS performance tests Type de document : Article/Communication Auteurs : H. Dai, Auteur ; L. Chen, Auteur ; C. Chen, Auteur Année de publication : 2012 Article en page(s) : pp 23 - 27 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Géomatique web [Termes IGN] analyse comparative [Termes IGN] architecture à trois niveaux [Termes IGN] diffusion de données [Termes IGN] serveur d'application [Termes IGN] test de performance [Termes IGN] traitement réparti [Termes IGN] WebSIG Résumé : (Auteur) A webGIS is usually constructed as a multi-tiers architecture basically consisting for the client side, the server side and the database component. From the client side, users send queries or requests to the server via (wireless) internet. Next the server communicates with database, where the requested information is compiled by the database server. Finally, the result is sent to the client. With the rapid growth of WebGIS application in 2010, the authors tested the performance of two models: centralized and distributed. Numéro de notice : A2012-006 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=31454 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
061-2012011	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Ecosystem survey data mapping interface, web-based GIS application for ocean species sample analysis / H. Dai in GIM international, vol 20 n° 2 (February 2006)

Public [article]  inGIM international > vol 20 n° 2 (February 2006) . - pp 61 - 63 Titre : Ecosystem survey data mapping interface, web-based GIS application for ocean species sample analysis Type de document : Article/Communication Auteurs : H. Dai, Auteur ; L. Chen, Auteur Année de publication : 2006 Article en page(s) : pp 61 - 63 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Applications SIG [Termes IGN] cartographie thématique [Termes IGN] écosystème [Termes IGN] faune aquatique [Termes IGN] interface web [Termes IGN] océanographie [Termes IGN] répartition géographique [Termes IGN] système d'information géographique Résumé : (Auteur) Data on ocean species has been being sampled along US coastlines for some sixty years. To enable scientists requiring such data for a variety of investigations to retrieve it and create dynamic spatial distribution GIS maps, the authors developed a Web-based application, the EcoSystem Survey Data Mapping Interface, which allows easy creation of dynamic GIS maps and interaction with them using GIS tools. Numéro de notice : A2006-139 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE Nature : Article DOI : sans Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=27866 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
061-06021	RAB	Revue	Centre de documentation	En réserve L003	Disponible

Nested hyper-rectangle learning model for remote sensing: land-cover classification / L. Chen in Photogrammetric Engineering & Remote Sensing, PERS, vol 71 n° 3 (March 2005)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 71 n° 3 (March 2005) . - pp 333 - 340 Titre : Nested hyper-rectangle learning model for remote sensing: land-cover classification Type de document : Article/Communication Auteurs : L. Chen, Auteur Année de publication : 2005 Article en page(s) : pp 333 - 340 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage dirigé [Termes IGN] classification dirigée [Termes IGN] classification par réseau neuronal [Termes IGN] image SPOT-HRV [Termes IGN] occupation du sol [Termes IGN] Taïwan Résumé : (Auteur) This study presents an exemplar-based nested hyper-rectangle learning model (NHLM) which is an efficient and accurate supervised classification model. The proposed model is based on the concept of seeding training data in the Euclidean m-space (where m denotes the number of features) as hyper-rectangles. To express the exceptions, these hyper-rectangles may be nested inside one another to an arbitrary depth. The fast and one-shot learning procedures can adjust weights dynamically when new examples are added. Furthermore, the "second chance" heuristic is introduced in NHLM to avoid creating more memory objects than necessary. NHLM is applied to solving the land cover classification problem in Taiwan using remote sensed imagery. The study investigated five land cover classes and clouds. These six classes were chosen from field investigation of the study area according to previous study. Therefore, this paper aims to produce a land cover classification based on SPOT HRV spectral data. Compared with a standard back-propagation neural network (BPN), the experimental results indicate that NHLM provides a powerful tool for categorizing remote sensing data. Numéro de notice : A2005-107 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.71.3.333 En ligne : https://doi.org/10.14358/PERS.71.3.333 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=27245 [article]

Merging R-trees: efficient strategies for local bulk insertion / L. Chen in Geoinformatica, vol 6 n° 1 (March - May 2002)  

Public [article]  inGeoinformatica > vol 6 n° 1 (March - May 2002) . - pp 7 - 34 Titre : Merging R-trees: efficient strategies for local bulk insertion Type de document : Article/Communication Auteurs : L. Chen, Auteur ; R. Choubey, Auteur ; E.A. Rundensteiner, Auteur Année de publication : 2002 Article en page(s) : pp 7 - 34 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Bases de données localisées [Termes IGN] arbre-R [Termes IGN] index spatial [Termes IGN] indexation spatiale [Termes IGN] intégration de données [Termes IGN] jeu de données localisées [Termes IGN] requête spatiale Résumé : (Auteur) A lot of recent work has focussed on bulk loading of data into multidimensional index structures in order to efficiently construct such structures for large data sets. In this paper, we address this problem with particular focus on R-trees - which are an important class of index structures used widely in commercial database systems. We propose a new technique, which as opposed to the current technique of inserting data one by one, bulk inserts entire new data sets into an active R-tree. This technique, called STLT (for small-tree-large-tree), considers the new data set as an R-tree itself (small tree), identifies and prepares a suitable location in the original R-tree (large tree) for insertion, and lastly performs the insert of the small tree into the large tree. Besides an analytical cost model of STLT, extensive experimental studies both on synthetic and real GIS data sets are also reported. These experiments not only compare STLT against the conventional technique, but also evaluate the suitability and limitations of STLT under different conditions, such as varying buffer sizes, ratio between existing and new data sizes, and skewness of new data with respect to the whole spatial region. We find that STLT does much better (in average, about 65%) than the existing technique for skewed data sets as well for large sizes of both the large tree and the small tree in terms of insertion time, while keeping comparable query tree quality. STLT consistently outperforms the alternate technique in all other circumstances in terms of bulk insertion time, especially, even up to 2,000% for the cases when the area of new data sets covers up to 4% of the global region covered by the existing index tree; however, at the cost of a deteriorating resulting tree quality. Numéro de notice : A2002-107 Affiliation des auteurs : non IGN Thématique : GEOMATIQUE/INFORMATIQUE Nature : Article DOI : 10.1023/A:1013764014000 En ligne : https://doi.org/10.1023/A:1013764014000 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=22022 [article]

Exemplaires(1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
057-02011	RAB	Revue	Centre de documentation	En réserve L003	Disponible

GBI : a generalized R-tree bulk-insertion strategy / R. Choubey (20/07/1999)  

Permalink