Centre de documentation
scientifique

Computationally efficient hyperspectral data learning based on the doubly stochastic dirichlet process / Xing Sun in IEEE Transactions on geoscience and remote sensing, vol 55 n° 1 (January 2017)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 55 n° 1 (January 2017) . - pp 363 - 374 Titre : Computationally efficient hyperspectral data learning based on the doubly stochastic dirichlet process Type de document : Article/Communication Auteurs : Xing Sun, Auteur ; Nelson H. C. Yung, Auteur ; Edmund Y. Lam, Auteur ; Hayden K.-H. So, Auteur Année de publication : 2017 Article en page(s) : pp 363 - 374 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] classification [Termes IGN] image hyperspectrale [Termes IGN] modèle stochastique [Termes IGN] problème de Dirichlet Résumé : (Auteur) The Dirichlet process (DP) prior is effective in modeling HSIs (HSI) and identifying land-cover classes. However, modeling a continuously varying intensity of these land covers elegantly and consistently is still a challenge. We propose a doubly stochastic DP (DSDP) as an efficient model of the global topic measurement space, which imposes a weaker assumption compared with the discrete Markov assumption, resulting in a lower computational cost than other DP-prior-based models. We also present a mixture model of DSDP, which is termed the marked sigmoidal Gaussian process (SGP) DSDP mixture model. It can be thinned from a DP mixture without massive auxiliary covariates, and the marked function prior makes the number of land-cover classes consistent, whereas the SGP function prior models the HSI land-cover variation globally. The consistency of the number of land covers is maintained for various HSIs with large-scale geographical areas. Experiments show that the model is robust and consistent on HSI identification with weak or even no supervision. Numéro de notice : A2017-020 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2016.2606575 En ligne : https://doi.org/10.1109/TGRS.2016.2606575 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=83951 [article]