Deep Autoencoder for Hyperspectral Unmixing via Global-Local Smoothing

Hyperspectral unmixing is to decompose the mixed pixels into pure spectral signatures (endmembers) and their proportions (abundances). Recently, deep learning-based methods have been applied to enhance the representation ability of unmixing models by extracting joint spatial-spectral characteristics...

Celý popis

Uloženo v:

Podrobná bibliografie
Vydáno v:	IEEE transactions on geoscience and remote sensing Ročník 60; s. 1 - 16
Hlavní autoři:	Xu, Xia, Song, Xinyu, Li, Tao, Shi, Zhenwei, Pan, Bin
Médium:	Journal Article
Jazyk:	angličtina
Vydáno:	New York IEEE 2022 The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:	Abundance Autoencoder (AE) Conditional random fields Convolution Data mining Deep learning Feature extraction global and local Homogeneity Hyperspectral imaging hyperspectral unmixing Imagery Methods Modules Neural networks Pixels Recurrent neural networks Risk reduction Self-similarity Smoothing Smoothing methods Spectra Spectral signatures Task analysis
ISSN:	0196-2892, 1558-0644
On-line přístup:	Získat plný text
Tagy:	Přidat tag Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!

Buďte první, kdo okomentuje tento záznam!