基于劈窗算法的地表温度反演算法

1.计算BAND10、BAND11亮温温度

（1）辐射标定

其中，Qcal为影像原始值

Landsat8参数列表：

    RADIANCE_MULT_BAND_1 = 1.2970E-02
    RADIANCE_MULT_BAND_2 = 1.3281E-02
    RADIANCE_MULT_BAND_3 = 1.2238E-02
    RADIANCE_MULT_BAND_4 = 1.0320E-02
    RADIANCE_MULT_BAND_5 = 6.3154E-03
    RADIANCE_MULT_BAND_6 = 1.5706E-03
    RADIANCE_MULT_BAND_7 = 5.2937E-04
    RADIANCE_MULT_BAND_8 = 1.1680E-02
    RADIANCE_MULT_BAND_9 = 2.4682E-03
    RADIANCE_MULT_BAND_10 = 3.3420E-04
    RADIANCE_MULT_BAND_11 = 3.3420E-04
    RADIANCE_ADD_BAND_1 = -64.84866
    RADIANCE_ADD_BAND_2 = -66.40582
    RADIANCE_ADD_BAND_3 = -61.19239
    RADIANCE_ADD_BAND_4 = -51.60087
    RADIANCE_ADD_BAND_5 = -31.57716
    RADIANCE_ADD_BAND_6 = -7.85295
    RADIANCE_ADD_BAND_7 = -2.64687
    RADIANCE_ADD_BAND_8 = -58.39799
    RADIANCE_ADD_BAND_9 = -12.34107
    RADIANCE_ADD_BAND_10 = 0.10000
    RADIANCE_ADD_BAND_11 = 0.10000

（2）亮温温度

公式：

2.计算大气水汽含量CWV

这部分没看懂，根据算法进行步骤总结
N应该是窗口大小，不知道有啥影响，好像是N越小越精确？
参考文献： A practical split-window algorithm for estimating land surface temperature from Landsat 8 data

3.计算地表比辐射率

（1）植被指数NDVI

NDVI=(NIR-RED)/(NIR+RED)
p.s. NIR:BAND4   RED:BAND5

（2）植被覆盖度FVC

FVC=(NDVI - NDVI_min) / (NDVI_max - NDVI_min)^2
p.s. NDVI_min=0.2    NDVI_max=0.5

（3）地表比辐射率LSE

4.计算地表温度

参考文献：Land Surface Temperature Retrieval from Landsat 8 TIRS—Comparison between Radiative Transfer Equation-Based Method, Split Window Algorithm and Single Channel Method

公式：

参数列表：