TY - CHAP
T1 - Introduction
T2 - Two fundamental principles behind hyperspectral imaging
AU - Chang, Chein I.
N1 - Publisher Copyright:
© 2023 John Wiley & Sons, Inc.
PY - 2022/11/11
Y1 - 2022/11/11
N2 - This chapter explores two fundamental principles, pigeon-hole principle and orthogonality principle, behind design and development of hyperspectral image processing algorithms in various applications. The pigeon-hole principle has found its applications in estimating the number of endmembers required by spectral unmixing, the number of spectrally distinct signatures by virtual dimensionality (VD), orders of low rank and sparse representation (LRaSR)/low rank and sparse matrix decomposition (LRaSMD), the number of bands to be selected for band selection (BS) and number of bands to be sampled for band sampling (BSam). The orthogonality principle can be used as a criterion to derive the orthogonal subspace projection (OSP) technique in developing algorithms for finding endmembers, detecting subpixel targets, anomalies as well as OSP-based LRaSMD in target/anomaly detection, and OSP-based detection theory applied to classification.
AB - This chapter explores two fundamental principles, pigeon-hole principle and orthogonality principle, behind design and development of hyperspectral image processing algorithms in various applications. The pigeon-hole principle has found its applications in estimating the number of endmembers required by spectral unmixing, the number of spectrally distinct signatures by virtual dimensionality (VD), orders of low rank and sparse representation (LRaSR)/low rank and sparse matrix decomposition (LRaSMD), the number of bands to be selected for band selection (BS) and number of bands to be sampled for band sampling (BSam). The orthogonality principle can be used as a criterion to derive the orthogonal subspace projection (OSP) technique in developing algorithms for finding endmembers, detecting subpixel targets, anomalies as well as OSP-based LRaSMD in target/anomaly detection, and OSP-based detection theory applied to classification.
UR - http://www.scopus.com/inward/record.url?scp=85147785425&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85147785425&partnerID=8YFLogxK
U2 - 10.1002/9781119687788.ch1
DO - 10.1002/9781119687788.ch1
M3 - Chapter
AN - SCOPUS:85147785425
SN - 9781119687788
SP - 3
EP - 40
BT - Advances in Hyperspectral Image Processing Techniques
PB - Wiley-Blackwell
ER -