TY - GEN
T1 - Hyperspectral Image Enhancement based on sensor simulation and vector decomposition
AU - Khandelwal, Ankush
AU - Rajan, K. S.
PY - 2011/9/13
Y1 - 2011/9/13
N2 - Hyperspectral Image Enhancement using multispectral data has received considerable attention in recent times in order to achieve higher classification accuracy and more detailed composition analysis. The objective is to obtain an image that has spectral resolution same as that of the hyperspectral image and spatial resolution same as that of multispectral image. While some of the fusion algorithms look at this as a band remapping problem, it is important to maintain the spectral band dependencies in such cases. In this paper, an attempt at using SRFs of different channels is presented to achieve hyperspectral and multispectral image fusion based on vector decomposition. Each multispectral channel fuses detail into only those hyperspectral channels which come into the sensitivity range of that multispectral channel. The results clearly show that the algorithm presented here successfully transfers the spatial details into hyperspectral data while maintaining spectral characteristics of that data.
AB - Hyperspectral Image Enhancement using multispectral data has received considerable attention in recent times in order to achieve higher classification accuracy and more detailed composition analysis. The objective is to obtain an image that has spectral resolution same as that of the hyperspectral image and spatial resolution same as that of multispectral image. While some of the fusion algorithms look at this as a band remapping problem, it is important to maintain the spectral band dependencies in such cases. In this paper, an attempt at using SRFs of different channels is presented to achieve hyperspectral and multispectral image fusion based on vector decomposition. Each multispectral channel fuses detail into only those hyperspectral channels which come into the sensitivity range of that multispectral channel. The results clearly show that the algorithm presented here successfully transfers the spatial details into hyperspectral data while maintaining spectral characteristics of that data.
KW - Hyperspectral Image Enhancement
KW - Multi-sensor data fusion
KW - Sensor simulation
KW - Spectral response functions
KW - Vector decomposition
UR - http://www.scopus.com/inward/record.url?scp=80052534303&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=80052534303&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:80052534303
SN - 9781457702679
T3 - Fusion 2011 - 14th International Conference on Information Fusion
BT - Fusion 2011 - 14th International Conference on Information Fusion
T2 - 14th International Conference on Information Fusion, Fusion 2011
Y2 - 5 July 2011 through 8 July 2011
ER -