Centre de documentation
scientifique

Learning to diversify deep belief networks for hyperspectral image classification / Ping Zhong in IEEE Transactions on geoscience and remote sensing, vol 55 n° 6 (June 2017)  

Public [article]  inIEEE Transactions on geoscience and remote sensing > vol 55 n° 6 (June 2017) . - pp 3516 - 3530 Titre : Learning to diversify deep belief networks for hyperspectral image classification Type de document : Article/Communication Auteurs : Ping Zhong, Auteur ; Zhiqiang Gong, Auteur ; Shutao Li, Auteur ; Carola-Bibiane Schönlieb, Auteur Année de publication : 2017 Article en page(s) : pp 3516 - 3530 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] apprentissage non-dirigé [Termes IGN] classification par réseau neuronal [Termes IGN] image hyperspectrale [Termes IGN] réseau neuronal convolutif [Termes IGN] théorie de Dempster-Shafer Résumé : (Auteur) In the literature of remote sensing, deep models with multiple layers have demonstrated their potentials in learning the abstract and invariant features for better representation and classification of hyperspectral images. The usual supervised deep models, such as convolutional neural networks, need a large number of labeled training samples to learn their model parameters. However, the real-world hyperspectral image classification task provides only a limited number of training samples. This paper adopts another popular deep model, i.e., deep belief networks (DBNs), to deal with this problem. The DBNs allow unsupervised pretraining over unlabeled samples at first and then a supervised fine-tuning over labeled samples. But the usual pretraining and fine-tuning method would make many hidden units in the learned DBNs tend to behave very similarly or perform as “dead” (never responding) or “potential over-tolerant” (always responding) latent factors. These results could negatively affect description ability and thus classification performance of DBNs. To further improve DBN’s performance, this paper develops a new diversified DBN through regularizing pretraining and fine-tuning procedures by a diversity promoting prior over latent factors. Moreover, the regularized pretraining and fine-tuning can be efficiently implemented through usual recursive greedy and back-propagation learning framework. The experiments over real-world hyperspectral images demonstrated that the diversity promoting prior in both pretraining and fine-tuning procedure lead to the learned DBNs with more diverse latent factors, which directly make the diversified DBNs obtain much better results than original DBNs and comparable or even better performances compared with other recent hyperspectral image classification methods. Numéro de notice : A2017-478 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article nature-HAL : ArtAvecCL-RevueIntern DOI : 10.1109/TGRS.2017.2675902 En ligne : http://dx.doi.org/10.1109/TGRS.2017.2675902 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=86403 [article]