Comparison of Two Spectral Mixture Analysis Models
WANG Qin-jun1, LIN Qi-zhong1, LI Ming-xiao2, WANG Li-ming1
1. Center for Earth Observation and Digital Earth, Chinese Academy of Sciences, Beijing 100101, China 2. China Earthquake Networks Center, Beijing 100045, China
Abstract：A spectral mixture analysis experiment was designed to compare the spectral unmixing effects of linear spectral mixture analysis (LSMA) and constraint linear spectral mixture analysis (CLSMA). In the experiment, red, green, blue and yellow colors were printed on a coarse album as four end members. Thirty nine mixed samples were made according to each end member’s different percent in one pixel. Then, field spectrometer was located on the top of the mixed samples’ center to measure spectrum one by one. Inversion percent of each end member in the pixel was extracted using LSMA and CLSMA models. Finally, normalized mean squared error was calculated between inversion and real percent to compare the two models’ effects on spectral unmixing. Results from experiment showed that the total error of LSMA was 0.300 87 and that of CLSMA was 0.375 52 when using all bands in the spectrum. Therefore, LSMA was 0.075 less than that of CLSMA when the whole bands of four end members’ spectra were used. On the other hand, the total error of LSMA was 0.280 95 and that of CLSMA was 0.298 05 after band selection. So, LSMA was 0.017 less than that of CLSMA when bands selection was performed. Therefore, whether all or selected bands were used, the accuracy of LSMA was better than that of CLSMA because during the process of spectrum measurement, errors caused by instrument or human were introduced into the model, leading to that the measured data could not mean the strict requirement of CLSMA and therefore reduced its accuracy. Furthermore, the total error of LSMA using selected bands was 0.02 less than that using the whole bands. The total error of CLSMA using selected bands was 0.077 less than that using the whole bands. So, in the same model, spectral unmixing using selected bands to reduce the correlation of end members’ spectra was superior to that using the whole bands.
Key words：Linear spectral mixture analysis;Constraint linear spectral mixture analysis;Remote sensing
王钦军1,蔺启忠1,黎明晓2,王黎明1 . 两种光谱混合分析模型的比较[J]. 光谱学与光谱分析, 2009, 29(10): 2602-2605.
WANG Qin-jun1, LIN Qi-zhong1, LI Ming-xiao2, WANG Li-ming1 . Comparison of Two Spectral Mixture Analysis Models. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2009, 29(10): 2602-2605.
[1] ZHOU Cun-lin, XU Han-qiu(周存林，徐涵秋). Journal of Image and Graphics(中国图象图形学报)，2007，12(5)：875. [2] LI Xiao-song, LI Zeng-yuan, WU Bo, et al(李晓松，李增元，吴 波，等). Journal of Remote Sensing(遥感学报)，2007，11(6)：923. [3] XU Jun, LI Ce, HUANG Xuan(许 珺，李 策，黄 绚). Remote Sensing Technology and Application(遥感技术与应用)，2000，15(1)：55. [4] Chen Xue-xia, LEE Vierling. Remote Sensing of Environment, 2006, 103: 338. [5] Miao Xin, Gong Peng, Swope Sarah, et al. Remote Sensing of Environment, 2006, 101: 329. [6] Zhang Jinkai, Rivard Benoit, Sa′nchez-Azofeifa Arturo. Remote Sensing of Environment, 2005, 95: 57. [7] Vikhamara Dagrun, Solberg Rune. Remote Sensing of Environment, 2003, 88: 309. [8] Sabol Jr. Donald E, Gillespie Alan R, Adams John B, et al. Remote Sensing of Environment, 2002, 80: 1. [9] Olthof Ian, Fraser Robert H. Remote Sensing of Environment, 2007, 107: 496. [10] Bannari A, Pacheco A, Staenz A, et al. Remote Sensing of Environment, 2006, 104: 447. [11] Fitzgerald Glenn J, Pinter Jr. Paul J, Hunsaker Douglas J, et al. Remote Sensing of Environment, 2005, 97: 526. [12] Eckmann Ted C, Roberts Dar A, Still Christopher J. Remote Sensing of Environment, 2008, 112: 3773. [13] Powell Rebecca L, Roberts Dar A, Dennison Philip E, et al. Remote Sensing of Environment, 2007, 106: 253. [14] TONG Heng-qing(童恒庆). Theoretical Econometrics(理论计量经济学). Beijing: Science Press(北京：科学出版社)，2005. 262.
