Centre de documentation
scientifique

Fusion of graph embedding and sparse representation for feature extraction and classification of hyperspectral imagery / Fulin Luo in Photogrammetric Engineering & Remote Sensing, PERS, vol 83 n° 1 (January 2017)  

Public [article]  inPhotogrammetric Engineering & Remote Sensing, PERS > vol 83 n° 1 (January 2017) . - pp 37 - 46 Titre : Fusion of graph embedding and sparse representation for feature extraction and classification of hyperspectral imagery Type de document : Article/Communication Auteurs : Fulin Luo, Auteur ; Hong Huang, Auteur ; Jiamin Liu, Auteur ; Zezhong Ma, Auteur Année de publication : 2017 Article en page(s) : pp 37 - 46 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] analyse discriminante [Termes IGN] classification [Termes IGN] extraction [Termes IGN] fusion de données multisource [Termes IGN] graphe [Termes IGN] image hyperspectrale Résumé : (Auteur) The graph embedding algorithms have been widely applied for feature extraction (FE) of hyperspectral imagery (HSI). These methods need to construct a similarity graph with k-nearest neighbors or ∈-radius ball. However, the neighborhood size is difficult to select in real-world applications. To solve the problem, we propose a new unsupervised FE method, called sparsity preserving analysis (SPA), based on sparse representation and graph embedding. The proposed algorithm utilizes sparse representation to obtain the sparse coefficients of data. Then, it constructs a new graph with the sparse coefficients that reveals the sparse properties of data. Finally, the structure of the graph is preserved in low-dimensional space to obtain a transformation matrix for FE. In addition, a new classification method, termed sparse neighborhood classifier (SNC), is designed using the sparse representation-based methodology. It uses the sparse coefficients of a new sample to obtain the similarity weights in each class. Then, the label information of the new sample is obtained by the weights. The classification accuracies of SPA with SNC reach to 86.9 percent and 80.6 percent on PaviaU and Urban HSI data sets, respectively. The results demonstrate that SPA with SNC can effectively extract low-dimensional features and improve the discriminating power for HSI classification. Numéro de notice : A2017-036 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Article DOI : 10.14358/PERS.83.1.37 En ligne : https://doi.org/10.14358/PERS.83.1.37 Format de la ressource électronique : URL article Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=84090 [article]