This algorithm is an implementation of the paper by Zhai et al, "Cloud/shadow detection based on spectral indices for multispectral optical remote sensing imagery", ISPRS Journal of Photogrammetry and Remote Sensing, volume 144, October 2018, Pages 235-253.
The authors recommend the following typical values for the parameters: SITS supports the following models:
t1: - (1,...,10) - default = 1
t2: - (0.1,...,0.5) - default = 0.11
t3: - (0.25,...,0.75) - default = 0.50
t4: - (0.5,...,0.90) - default = 0.75
t5: - (30,...,90) - default = 40
t6: - (3,...,11) - default = 5
Please see Zhai et al.'s paper for more detail.
sits_cloud_cbers(
cube,
cld_band_name = "CMASK",
data_dir = NULL,
t1 = 1,
t2 = 0.11,
t3 = 0.5,
t4 = 0.75,
t5 = 40,
t6 = 5,
memsize = 8,
multicores = 2
)input data cube
indication of the cloud band to be produced
directory where cloud band will be written
controls the difference btw visible and infrared bands
controls the brightness properties of cloud
controls the dark property of cloud shadows.
remove influence of water in the cloud shadow detection
size of window to search for clouds near shadows
size of window of median filter to remove outliers
size of memory
number of cores
new data cube with cloud data