def mask_object(cube, decimate=False, remove_nan=False, **kwargs):
obj_rmin = kwargs.get('obj_rmin', None)
obj_rmax = kwargs.get('obj_rmax', None)
if obj_rmin is not None or obj_rmax is not None:
obj_mask = solar.define_map_mask(cube, **kwargs)
# decimate is mandatory to remove nan because NaN * 0 = NaN
if decimate or remove_nan:
Mo = lo.decimate(obj_mask, dtype=cube.dtype)
else:
Mo = lo.ndmask(obj_mask, dtype=cube.dtype)
return Mo, obj_mask
def stsrt(data, cube, **kwargs):
"""
Smooth Temporal Solar Rotational Tomography.
Assumes data is sorted by observation time 'DATE_OBS'.
Returns
-------
P : The projector with masking
D : Smoothness priors
obj_mask : object mask array
data_mask : data mask array
"""
# Parse kwargs.
obj_rmin = kwargs.get('obj_rmin', None)
obj_rmax = kwargs.get('obj_rmax', None)
# mask data
data_mask = solar.define_data_mask(data, **kwargs)
# define temporal groups
times = [solar.convert_time(h['DATE_OBS']) for h in data.header]
## if no interval is given separate every image
dt_min = kwargs.get('dt_min', np.max(np.diff(times)) + 1)
#groups = solar.temporal_groups(data, dt_min)
ind = solar.temporal_groups_indexes(data, dt_min)
n = len(ind)
# define new 4D cube
cube4 = cube[..., np.newaxis].repeat(n, axis=-1)
cube4.header = copy.copy(cube.header)
cube4.header['NAXIS'] = 4
cube4.header['NAXIS4'] = cube4.shape[3]
# define 4d model
# XXX assumes all groups have same number of elements
ng = data.shape[-1] / n
P = siddon4d_lo(data.header, cube4.header, ng=ng, mask=data_mask, obstacle="sun")
# priors
D = smoothness_prior(cube4, kwargs.get("height_prior", False))
# mask object
if obj_rmin is not None or obj_rmax is not None:
Mo, obj_mask = mask_object(cube, **kwargs)
obj_mask = obj_mask[..., np.newaxis].repeat(n, axis=-1)
if kwargs.get("decimate", False) or kwargs.get("remove_nan", False):
Mo = lo.decimate(obj_mask)
else:
Mo = lo.ndmask(obj_mask)
P = P * Mo.T
D = [Di * Mo.T for Di in D]
else:
obj_mask = None
return P, D, obj_mask, data_mask
