Utilize este identificador para referenciar este registo: http://hdl.handle.net/10400.21/3611
Título: Dependent component analysis: a hyperspectral unmixing algorithm
Autor: Nascimento, José M. P.
Bioucas-Dias, José M.
Palavras-chave: Hyperspectral unmixing algorithm
Data: Jun-2007
Editora: Springer Berlin Heidelberg
Citação: NASCIMENTO, José M. P.; BIOUCAS-DIAS, José M. - Dependent Component Analysis: A Hyperspectral Unmixing Algorithm. Pattern Recognition and Image Analysis. Vol. 4478 (2007), p. 612-619.
Relatório da Série N.º: Lecture Notes in Computer Science;
Resumo: Linear unmixing decomposes a hyperspectral image into a collection of reflectance spectra of the materials present in the scene, called endmember signatures, and the corresponding abundance fractions at each pixel in a spatial area of interest. This paper introduces a new unmixing method, called Dependent Component Analysis (DECA), which overcomes the limitations of unmixing methods based on Independent Component Analysis (ICA) and on geometrical properties of hyperspectral data. DECA models the abundance fractions as mixtures of Dirichlet densities, thus enforcing the constraints on abundance fractions imposed by the acquisition process, namely non-negativity and constant sum. The mixing matrix is inferred by a generalized expectation-maximization (GEM) type algorithm. The performance of the method is illustrated using simulated and real data.
Peer review: yes
URI: http://hdl.handle.net/10400.21/3611
ISBN: 978-3-540-72848-1
978-3-540-72849-8
Versão do Editor: http://link.springer.com/chapter/10.1007%2F978-3-540-72849-8_77
Aparece nas colecções:ISEL - Eng. Elect. Tel. Comp. - Comunicações

Ficheiros deste registo:
Ficheiro Descrição TamanhoFormato 
Dependent Component Analysis A Hyperspectral Unmixing Algorithm.pdf964,7 kBAdobe PDFVer/Abrir    Acesso Restrito. Solicitar cópia ao autor!


FacebookTwitterDeliciousLinkedInDiggGoogle BookmarksMySpace
Formato BibTex MendeleyEndnote Degois 

Todos os registos no repositório estão protegidos por leis de copyright, com todos os direitos reservados.