Abstract | ||
---|---|---|
Matrix factorization is applied to unsupervised linear unmixing for hyperspectral imagery. The algorithm, called non-negative matrix factorization, is used. It imposes a constraint on the non-negativity of the amplitudes of the recovered endmember spectral signatures as well as their fractional abundances. This ensures the recovery of physically meaningful endmembers and their abundances. This algorithm is further modified to include the sum-to-one constraint such that the sum of the fractional abundances for each pixel is unity. Several practical implementation issues in hyperspectral image unmixing are discussed. Some preliminary results from AVERIS experiments are presented. We also discuss the advantages and possible limitations of this method in hyperspectral image analysis. |
Year | DOI | Venue |
---|---|---|
2005 | 10.1109/IGARSS.2005.1525870 | IGARSS |
Keywords | Field | DocType |
nonnegative matrix factorization,independent component analysis,non negative matrix factorization,image processing,algorithms,hyperspectral imaging,mathematical models,remote sensing,least squares approximation,image analysis,layout,matrix factorization,pixels,hyperspectral imagery,hyperspectral sensors,vectors,principal component analysis,pixel | Endmember,Computer science,Remote sensing,Image processing,Artificial intelligence,Computer vision,Pattern recognition,Matrix decomposition,Multispectral image,Hyperspectral imaging,Pixel,Non-negative matrix factorization,Image resolution | Conference |
Volume | ISBN | Citations |
6 | 0-7803-9050-4 | 1 |
PageRank | References | Authors |
0.37 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qian Du | 1 | 2833 | 185.90 |
Ivica Kopriva | 2 | 146 | 16.60 |
Harold Szu | 3 | 149 | 38.33 |