Title :
Total variation based hyperspectral feature extraction
Author :
Rasti, Behnood ; Sveinsson, Johannes R. ; Ulfarsson, Magnus O.
Author_Institution :
Fac. of Electr. & Comput. Eng., Univ. of Iceland, Reykjavik, Iceland
Abstract :
In this paper, a hyperspectral feature extraction method is proposed. A low-rank linear model using the right eigenvector of the observed data is given for hyperspectral images. A total variation (TV) based regularization called Low-Rank TV regularization (LRTV) is used for hyperspectral feature extraction. The feature extraction is used for hyperspectral image classification. The classification accuracies obtained are significantly better than the ones obtained using features extracted by Principal Component Analysis (PCA) and Maximum Noise Fraction (MNF).
Keywords :
eigenvalues and eigenfunctions; feature extraction; geophysical image processing; hyperspectral imaging; image classification; LRTV; eigenvector; hyperspectral image classification; hyperspectral images; low rank TV regularization; low rank linear model; total variation based hyperspectral feature extraction; total variation based regularization; Feature extraction; Hyperspectral imaging; Noise; Principal component analysis; Radio frequency; Support vector machines; Feature extraction; hyperspectral image; low-rank model; regularization; total variation;
Conference_Titel :
Geoscience and Remote Sensing Symposium (IGARSS), 2014 IEEE International
Conference_Location :
Quebec City, QC
DOI :
10.1109/IGARSS.2014.6947528
