def satMask(location):
print("\n-> Building mask of saturated stars...")
masks = []
images = glob.glob("%s/data/*.fits" % (location))
for i in images:
hdu = fits.open(i, mode='update')
try:
hdu[0].header['MASKED']
except KeyError:
(hdu[0].header).set('MASKED', 'N')
if hdu[0].header['MASKED'] == 'N':
sat = hdu[0].header['SATURATE']
data = np.empty((hdu[0].data).shape, dtype=np.float32)
data[:,:] = (hdu[0].data)[:,:]
emptyMask = np.zeros(data.shape, dtype=np.uint8, order='C')
astroscrappy.update_mask(np.asarray(data), np.asarray(emptyMask), sat, True)
masks.append(emptyMask)
hdu.close()
if masks != []:
and_comb = np.zeros(masks[0].shape)
for x in range(masks[0].shape[0]):
for y in range(masks[0].shape[1]):
pixels = []
for m in range(len(masks)):
pixels.append(masks[m][x,y])
truth = int(np.all(pixels))
and_comb[x,y] = truth
return and_comb.astype(int)
def sat_test(image):
#get saturation mask and combine with bad pixel mask
data = fits.getdata(image)
bpm = fits.getdata(image, 1)
sat = fits.getval(image, 'SATURATE')
data_sat = np.empty((data).shape, dtype=np.float32)
data_sat[:,:] = (data)[:,:]
emptyMask = np.zeros(data_sat.shape, dtype=np.uint8, order='C')
astroscrappy.update_mask(np.asarray(data_sat), np.asarray(emptyMask), sat, True)
return bpm, emptyMask, sat
def cosmic(image, objLim=5.0, mask_ext=1, gain_key='GAIN', readnoise_key='RDNOISE',
sat_key='SATURATE', lin_key='MAXLIN'):
hdu = fits.open(image, mode='update')
hdr = hdu[0].header
try:
hdr['MASKED']
except KeyError:
hdr.set('MASKED', 'N')
hdu.close()
hdu = fits.open(image, mode='update')
hdr = hdu[0].header
if hdr['MASKED'] == 'N':
data = hdu[0].data
nx, ny = data.shape[0], data.shape[1]
try:
bpm = hdu['BPM'].data
except KeyError:
try:
bpm = fits.getdata(image, mask_ext)
bpm = bpm.astype(int)
except Exception as esc:
print(esc)
bpm = np.zeros((nx,ny))
bpm_bool = bpm.astype(bool)
Gain = hdr[gain_key]
readNoise = hdr[readnoise_key]
sat = hdr[sat_key]
if lin_key != 'NONE':
lin = hdr[lin_key]
else:
lin = sat
if lin < sat:
sat = float(lin)
#get saturation mask and combine with bad pixel mask
data_sat = np.empty((hdu[0].data).shape, dtype=np.float32)
data_sat[:,:] = (hdu[0].data)[:,:]
emptyMask = np.zeros(data_sat.shape, dtype=np.uint8, order='C')
astroscrappy.update_mask(np.asarray(data_sat), np.asarray(emptyMask), sat, True)
bpm = np.logical_or(bpm, emptyMask)
#mask cosmic rays
cosmicMask, cosmicData = astroscrappy.detect_cosmics(data, inmask=bpm_bool,
objlim=objLim, gain=Gain,
readnoise=readNoise,
satlevel=sat)
cosmicMask = cosmicMask.astype(np.int8)
#combine BPM and cosmic ray mask
badPixelMask = np.empty(bpm.shape, dtype=np.uint8, order='C')
badPixelMask[:,:] = bpm[:,:]
masterMask = np.asarray(badPixelMask) | np.asarray(cosmicMask)
hdr.set('MASKED', 'Y')
hduData = fits.PrimaryHDU(cosmicData, header=hdr)
hduMask = fits.ImageHDU(masterMask)
hduList= fits.HDUList([hduData, hduMask])
hduList.writeto(image, overwrite=True)
hdu.close()
def create_mask(science_file, red, suffix, static_mask, source_mask,
saturation, binning, rdnoise, sigclip, sigfrac, objlim, log):
t_start = time.time()
log.info('Running create_mask version: ' + str(__version__))
with fits.open(static_mask) as hdr:
mask_bp = -~-hdr[0].data
data = np.ascontiguousarray(red.data.astype(np.float32))
log.info('Masking saturated pixels.')
mask_sat = np.zeros(
(np.shape(data)[0],
np.shape(data)[1])).astype(np.bool) #create empty mask
astroscrappy.update_mask(data, mask_sat, saturation,
True) #add saturated stars to mask
mask_sat = mask_sat.astype(np.uint8) #set saturated star mask type
mask_sat[data >= saturation] = 4 #set saturated pixel flag
mask_sat[mask_sat == 1] = 8 #set connected to saturated pixel flag
log.info('Cleaning and masking cosmic rays.')
log.info('Using sigclip = %d, sigfrac = %.1f, objlim = %d' %
(sigclip, sigfrac, objlim))
mask_cr, clean = astroscrappy.detect_cosmics(
data,
inmask=(mask_bp + mask_sat + source_mask).astype(np.bool),
sigclip=sigclip,
sigfrac=sigfrac,
readnoise=rdnoise,
satlevel=saturation,
objlim=objlim
) #clean science image of cosmic rays and create cosmic ray mask
mask_cr = mask_cr.astype(np.uint8) #set cosmic ray mask type
mask_cr[mask_cr == 1] = 2 #set cosmic ray flag
log.info('Maksing satellite trails.')
satellite_fitting = False
red_path = os.path.dirname(science_file.replace('/raw/', '/red/'))
binned_data = clean.reshape(int(np.shape(clean)[0] / binning[0]),
binning[0],
int(np.shape(clean)[1] / binning[1]),
binning[1]).sum(3).sum(1) #bin data
for j in range(3):
fits.PrimaryHDU(binned_data).writeto(
red_path + '/binned_mask.fits',
overwrite=True) #write binned data to tmp file
results, errors = detsat(red_path + '/binned_mask.fits',
chips=[0],
n_processes=1,
buf=40,
sigma=3,
h_thresh=0.2) #detect sateliite trails
trail_coords = results[(red_path + '/binned_mask.fits',
0)] #create satellite trail if found
if len(trail_coords) > 0: #continue if sateliite trail found
trail_segment = trail_coords[0]
try:
mask_binned = make_mask(
red_path + '/binned_mask.fits',
0,
trail_segment,
sublen=5,
pad=0,
sigma=5,
subwidth=5000).astype(
np.uint8) #create satellite trail mask
except ValueError:
log.error('Satellite trail could not be fitted for file ' +
science_file + ' and is not included in the mask. '
) #if error occurs, add comment
break
satellite_fitting = True
binned_data[mask_binned == 1] = np.median(binned_data)
try:
open_old_mask = fits.open(red_path + '/old_mask.fits')
old_mask = open_old_mask[0].data
open_old_mask.close()
mask_binned = old_mask + mask_binned
except IOError:
pass
fits.writeto(red_path + '/old_mask.fits',
mask_binned,
overwrite=True)
else:
break
os.remove(red_path + '/binned_mask.fits')
if os.path.exists(red_path + '/old_mask.fits'):
os.remove(red_path + '/old_mask.fits')
if satellite_fitting == True:
mask_sate = np.kron(mask_binned, np.ones(
(binning[0], binning[1]))).astype(np.uint8) #unbin mask
mask_sate[mask_sate == 1] = 16 #set satellite trail flag
else: #if no satellite trails are found, create empty mask
mask_sate = (np.zeros([np.shape(clean)[0],
np.shape(clean)[1]])).astype(np.uint8)
mask = mask_bp + mask_cr + mask_sat + mask_sate #combine bad pixel, cosmic ray, saturated star and satellite trail masks
mask_hdu = fits.PrimaryHDU(mask) #create mask Primary HDU
mask_hdu.header['VER'] = (__version__, 'Version of create mask used.'
) #name of log file
mask_hdu.header['USE'] = ('Complex mask using additive flags.',
'e.g. 6 = 2 + 4') #header comment
mask_hdu.header['M-BP'] = (1, 'Value of masked bad pixels.')
mask_hdu.header['M-BPNUM'] = (np.sum(mask & 1 == 1),
'Number of bad pixels.')
mask_hdu.header['M-CR'] = (2, 'Value of masked cosmic ray pixels.')
mask_hdu.header['M-CRNUM'] = (np.sum(mask & 2 == 2),
'Number of cosmic ray pixels.')
mask_hdu.header['SATURATE'] = (saturation, 'Level of saturation.')
mask_hdu.header['M-SP'] = (4, 'Value of masked saturated pixels.')
mask_hdu.header['M-SPNUM'] = (np.sum(mask & 4 == 4),
'Number of saturated pixels.')
mask_hdu.header['M-CSP'] = (8,
'Value of masked saturated-connected pixels.')
mask_hdu.header['M-CSPNUM'] = (np.sum(mask & 8 == 8),
'Number of saturated-connected pixels.')
mask_hdu.header['M-SAT'] = (16, 'Value of masked satellite trail pixels.')
mask_hdu.header['M-SATNUM'] = (np.sum(mask & 16 == 16),
'Number of satellite trail pixels.')
mask_hdu.writeto(science_file.replace('/raw/',
'/red/').replace('.fits', suffix),
overwrite=True) #write mask to file
log.info('Mask created.')
t_end = time.time()
log.info('Mask creation completed in ' + str(t_end - t_start) + ' sec')
return clean, mask