Fully constrained least-squares based linear unmixing

Daniel Heinz; Chein I. Chang; Mark L.G. Althouse

Fully constrained least-squares based linear unmixing

Daniel Heinz
, Chein I. Chang
, Mark L.G. Althouse

Department of Electrical Engineering

Research output: Contribution to conference › Paper › peer-review

116 Citations (Scopus)

Abstract

A fully constrained least-squares linear unmixing approach to hyperspectral image classification is presented. It is derived from an unconstrained least-squares based orthogonal subspace projection. It is similar to a method developed by Shimabukuro and Smith in the least-squares sense, but significantly different from their method in the way of implementing the constraints. Since there is no closed form solution available, an efficient algorithm is developed for finding a fully constrained solution, which can be viewed as a generalization of Shimabukuro and Smith's method. The effectiveness of this algorithm is demonstrated through computer simulations and real data experiments.

Original language	English
Pages	1401-1403
Number of pages	3
Publication status	Published - 1999
Event	Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century' - Hamburg, Ger Duration: 1999 Jun 28 → 1999 Jul 2

Conference

Conference	Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century'
City	Hamburg, Ger
Period	99-06-28 → 99-07-02

All Science Journal Classification (ASJC) codes

Computer Science Applications
General Earth and Planetary Sciences

Cite this

@conference{94573dfb468840019729eb1356b706bb,

title = "Fully constrained least-squares based linear unmixing",

abstract = "A fully constrained least-squares linear unmixing approach to hyperspectral image classification is presented. It is derived from an unconstrained least-squares based orthogonal subspace projection. It is similar to a method developed by Shimabukuro and Smith in the least-squares sense, but significantly different from their method in the way of implementing the constraints. Since there is no closed form solution available, an efficient algorithm is developed for finding a fully constrained solution, which can be viewed as a generalization of Shimabukuro and Smith's method. The effectiveness of this algorithm is demonstrated through computer simulations and real data experiments.",

author = "Daniel Heinz and Chang, \{Chein I.\} and Althouse, \{Mark L.G.\}",

year = "1999",

language = "English",

pages = "1401--1403",

note = "Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century' ; Conference date: 28-06-1999 Through 02-07-1999",

}

Fully constrained least-squares based linear unmixing. / Heinz, Daniel; Chang, Chein I.; Althouse, Mark L.G.
1999. 1401-1403 Paper presented at Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century', Hamburg, Ger.

Research output: Contribution to conference › Paper › peer-review

TY - CONF

T1 - Fully constrained least-squares based linear unmixing

AU - Heinz, Daniel

AU - Chang, Chein I.

AU - Althouse, Mark L.G.

PY - 1999

Y1 - 1999

N2 - A fully constrained least-squares linear unmixing approach to hyperspectral image classification is presented. It is derived from an unconstrained least-squares based orthogonal subspace projection. It is similar to a method developed by Shimabukuro and Smith in the least-squares sense, but significantly different from their method in the way of implementing the constraints. Since there is no closed form solution available, an efficient algorithm is developed for finding a fully constrained solution, which can be viewed as a generalization of Shimabukuro and Smith's method. The effectiveness of this algorithm is demonstrated through computer simulations and real data experiments.

AB - A fully constrained least-squares linear unmixing approach to hyperspectral image classification is presented. It is derived from an unconstrained least-squares based orthogonal subspace projection. It is similar to a method developed by Shimabukuro and Smith in the least-squares sense, but significantly different from their method in the way of implementing the constraints. Since there is no closed form solution available, an efficient algorithm is developed for finding a fully constrained solution, which can be viewed as a generalization of Shimabukuro and Smith's method. The effectiveness of this algorithm is demonstrated through computer simulations and real data experiments.

UR - https://www.scopus.com/pages/publications/0033326640

UR - https://www.scopus.com/pages/publications/0033326640#tab=citedBy

M3 - Paper

AN - SCOPUS:0033326640

SP - 1401

EP - 1403

T2 - Proceedings of the 1999 IEEE International Geoscience and Remote Sensing Symposium (IGARSS'99) 'Remote Sensing of the Systems Earth - A Challenge for the 21st Century'

Y2 - 28 June 1999 through 2 July 1999

ER -

Fully constrained least-squares based linear unmixing

Abstract

Conference

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this