Title | ||
---|---|---|
A Subspace-Based Multinomial Logistic Regression for Hyperspectral Image Classification |
Abstract | ||
---|---|---|
In this letter, we propose a multinomial-logistic-regression method for pixelwise hyperspectral classification. The feature vectors are formed by the energy of the spectral vectors projected on class-indexed subspaces. In this way, we model not only the linear mixing process that is often present in the hyperspectral measurement process but also the nonlinearities that are separable in the feature space defined by the aforementioned feature vectors. Our experimental results have been conducted using both simulated and real hyperspectral data sets, which are collected using NASA's Airborne Visible/Infrared Imaging Spectrometer (AVIRIS) and the Reflective Optics System Imaging Spectrographic (ROSIS) system. These results indicate that the proposed method provides competitive results in comparison with other state-of-the-art approaches. |
Year | DOI | Venue |
---|---|---|
2014 | 10.1109/LGRS.2014.2320258 | Geoscience and Remote Sensing Letters, IEEE |
Keywords | Field | DocType |
geophysical image processing,hyperspectral imaging,image classification,mixing,regression analysis,vectors,AVIRIS,NASA Airborne Visible-Infrared Imaging Spectrometer,ROSIS system,Reflective Optics System Imaging Spectrographic system,class-indexed subspace,feature vector formation,hyperspectral measurement process,linear mixing processing,pixelwise hyperspectral image classification,spectral vector energy,subspace-based multinomial logistic regression method,Hyperspectral imaging,pixelwise classification,subspace multinomial logistic regression (MLR) | Data set,Remote sensing,Separable space,Artificial intelligence,Computer vision,Imaging spectrometer,Feature vector,Full spectral imaging,Subspace topology,Pattern recognition,Linear subspace,Hyperspectral imaging,Mathematics | Journal |
Volume | Issue | ISSN |
11 | 12 | 1545-598X |
Citations | PageRank | References |
10 | 0.53 | 8 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mahdi Khodadadzadeh | 1 | 68 | 9.12 |
Jun Li 0009 | 2 | 10 | 0.53 |
Antonio Plaza | 3 | 3475 | 262.63 |
José M. Bioucas-Dias | 4 | 3565 | 173.67 |