An Iterative Search in End-Member Fraction Space for Spectral Unmixing

• Published in: IEEE geoscience and remote sensing letters Vol. 8; no. 4; pp. 706 - 709
• Main Authors: Shoshany, M., Kizel, F., Netanyahu, N. S., Goldshlager, N., Jarmer, T., Even-Tzur, G.
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 01.07.2011; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Approximation algorithms; Assessments; EMS; Hyperspectral imagery; Image reconstruction; Iterative methods; Least squares method; Materials; Mathematical models; Pixel; Search problems; Searching; Signal to noise ratio; Soil; Spectra; unmixing

A novel unmixing methodology is presented, searching for a fraction combination of end-members (EMs) that reconstructs the integrated source signal. The search starts with computing an initially estimated unmixing solution and then assesses combinations selected at random within an envelope surround...