Denggang Li, Zhongmei Wang. Improved Spatial Information Constrained Nonnegative Matrix Factorization Method for Hyperspectral Unmixing[J]. Laser & Optoelectronics Progress, 2019, 56(11): 111006
- Laser & Optoelectronics Progress
- Vol. 56, Issue 11, 111006 (2019)
Fig. 1. Quadratic parabola (σ=4, r=2)
Fig. 2. Endmembers selected from USGS spectral library
Fig. 3. RSAD and RRMSE versus β. (a) RSAD; (b) RRMSE
Fig. 4. RSAD and RRMSE versus α. (a) RSAD; (b) RRMSE
Fig. 5. RSAD and RRMSE versus k. (a) RSAD; (b) RRMSE
Fig. 6. RSAD and RRMSE versus σ. (a) RSAD; (b) RRMSE
Fig. 7. RSAD and RRMSE versus RSNR. (a) RSAD; (b) RRMSE
Fig. 8. Abundance maps after unmixing by ASSNMF method. (a) Roof; (b) tree; (c) meadow; (d) pitch
|
Table 1. RSAD with and without pure pixels
|
Table 2. RRMSE with and without pure pixels
|
Table 3. RSAD of real spectra and endmember spectra extracted by each algorithm
Set citation alerts for the article
Please enter your email address
© Copyright 2018-2021 | Chinese Laser Press. All Rights Reserved 沪ICP备15018463号-20