The CM_PixMask Module

The CM_PixMask module defines the CMPixMask class, which implements a pixel-based masking scheme for specifying which regions of a given satellite image to process.

The CMPixMask class

The CMPixMask can be initialised with four masks, each of which should use a value of 0 to indicate a pixel to be processed and a value of greater than zero for a pixel to mask. The four masks are nominally defined as:

• cloud_mask: An array containing pixels to be masked due to cloud contamination.

• land_mask: An array containing pixels to be masked as they are not water.

• sol_zen_mask: An array containing pixels to be masked due to their high solar zenith angle.

• sat_zen_mask: An array containing pixels to be masked due to their high satellite zenith angle.

These four masks are optional, and any - or none - can be specified. The core of CMPixMask, once initialised, is the CMPixMask.mask array, which contains a combined mask utilising all of the other masks, if supplied. If no masks are supplied, all pixels are processed.

The cut_high_zen helper function

The CMPixMask class also contains a helper function, cut_high_zen that can compute a mask based on a user-supplied zenith angle array and threshold value:

from pycoxmunk.CM_PixMask import CMPixMask
cm_mask = CMPixMask()
my_solar_zeniths = load_solzens()  # where this is some function to load zenith angles
cm_mask.cut_high_zen(my_solar_zeniths, threshold=70)

The above code will set all pixels in cm_mask.mask to 1 where my_solar_zeniths is greater than 70°. Be aware of confusion between degrees and radians, the function is agnostic but requires both to be in the same units. Typically, zenith (and azimuth) angles for satellites are specified in degrees, but this is not always the case.