def main():
folder = os.environ['HOME']+'/data/hirise/inca'
# dem = ImgData(os.path.join(folder, 'big_spider_dem.cub'))
slopes = ImgData(os.path.join(folder,'big_spider_slopes.tif'))
aspects = ImgData(os.path.join(folder,'big_spider_aspects.tif'))
spider1 = ImgData(os.path.join(folder,'ESP_022607_0985_cropped_big_spider.cub'))
# spider2 = ImgData(os.path.join(folder,'ESP_022699_0985_cropped_big_spider.cub'))
# read in required data
# dem.read_all()
slopes.read_cropped_by_n(1)
aspects.read_cropped_by_n(1)
spider1.read_all()
# correct_azimuth(dem, aspects)
# clean up data
slopes.data[slopes.data < 0] = np.nan
slopes.data[slopes.data > 90] = np.nan
aspects.data[aspects.data < 0] = np.nan
aspects.data[aspects.data > 360] = np.nan
# create MarsSpicer object
# mspice = MarsSpicer()
# get labels from observation
labels = pdstools.get_labels('/Users/maye/data/hirise/inca/ESP_022607_0985_RED.LBL')
# mspice.utc = pdstools.get_time(labels)
#set coordinates
# mspice.set_spoint_by(lat=pdstools.get_mean_lat(labels),lon=pdstools.get_mean_lon(labels))
label_incidence = pdstools.get_incidence(labels)
label_sol_azi = pdstools.get_sub_sol_azi(labels)
# correct for difference to GDAL coordinates (not correcting the aspects map)
sol_azi = 90 + label_sol_azi
# using negative sol_azi because angles are positive in clock-direction in HiRISE data
sun_vec = get_xyz(label_incidence, -sol_azi)
# now the S/C (emission) vector
sc_azimuth = pdstools.get_sub_sc_azimuth(labels)
# shift to GDAL 0-point, as above
sc_azimuth += 90
emis = pdstools.get_emission(labels)
# again, use negative azimuth, as the formula for xyz uses positive counter-clockwise
sc_vec = get_xyz(emis, -sc_azimuth)
print("slopes min max: {0}, {1}".format(np.nanmin(slopes.data),np.nanmax(slopes.data)))
print("aspect min max: {0}, {1}".format(np.nanmin(aspects.data),np.nanmax(aspects.data)))
print(slopes.data.argmin(),slopes.data.argmax())
print(aspects.data.argmin(),aspects.data.argmax())
normal_xyz = get_xyz(slopes.data, 360-aspects.data)
normal_x,normal_y,normal_z = normal_xyz
normal_vec = np.dstack([normal_x,normal_y,normal_z])
# first calculate real incidences
cosa = normal_vec.dot(sun_vec) / np.apply_along_axis(np.linalg.norm,2,normal_vec)
cosa /= np.linalg.norm(sun_vec)
print('cosa calculated.')
# now real emissions
cose = normal_vec.dot(sc_vec) / np.apply_along_axis(np.linalg.norm,2,normal_vec)
cose /= np.linalg.norm(sc_vec)
print('cose calculated.')
# set smaller than 0 to 0
cosa[cosa<0.0] = 0.0
real_inc = np.degrees(np.arccos(cosa))
real_emis = np.degrees(np.arccos(cose))
# if current_inc >= 89.0:
# cos_corrected[line,sample] = np.nan
# else:
# value = math.cos(math.radians(current_inc))
# cos_corrected[line,sample]= value
cos_corrected = cosa
print('min/max:{0}, {1}'.format(np.nanmin(real_inc),np.nanmax(real_inc)))
print('min/max:{0}, {1}'.format(np.nanmin(cos_corrected),np.nanmax(cos_corrected)))
print('min/max:{0}, {1}'.format(np.nanmin(cose),np.nanmax(cose)))
print(np.histogram(cos_corrected,range=(0,1)))
print(np.histogram(real_inc,range=(0,90)))
fig = plt.figure()
ax_inc = fig.add_subplot(122)
# ax_inc = fig.add_subplot(221)
img = ax_inc.imshow(cos_corrected,cmap = 'gray')
ax_inc.set_title("'Lambertian' image from Inca DEM",fontsize=14)
# colorbar(img)
plt.axis('off')
ax_real = fig.add_subplot(121)
img = ax_real.imshow(spider1.data,cmap='gray')
ax_real.set_title('Ortho-image Inca spider',fontsize=14)
# ax_emis = fig.add_subplot(222)
# img = ax_emis.imshow(cose, cmap='gray')
# colorbar(img)
# ax_emis.set_title('cose')
# ax_hist = fig.add_subplot(223)
# emis_angles = np.degrees(np.arccos(cose))
# ax_hist.hist(emis_angles.flatten(),bins=45,range=(0,90))
# ax_hist.set_title('Emission angles histogram')
plt.axis('off')
plt.show()
return [real_inc,real_emis]
delta_angle = np.degrees(np.arctan2(v1[1],v1[0]))-90.0
dsample = newPoint.sample - dem.center.sample
dline = newPoint.line - dem.center.line
# quiver(dem.center.sample,dem.center.line,dsample,dline,angles='xy', scale_units='xy', scale=1)
# correct aspects for delta angle
# it needs to be added, because aspects go clock-wise
# DANGER: This only works
aspects.data += delta_angle
# bend around data > 360
mask = aspects.data > 360.0
aspects.data[mask] = aspects.data[mask] - 360.0
folder = os.environ['HOME']+'/data/hirise/inca'
dem = ImgData(os.path.join(folder, 'big_spider_dem.cub'))
slopes = ImgData(os.path.join(folder,'big_spider_slopes.tif'))
aspects = ImgData(os.path.join(folder,'big_spider_aspects.tif'))
spider1 = ImgData(os.path.join(folder,'ESP_022607_0985_cropped_big_spider.cub'))
# spider2 = ImgData(os.path.join(folder,'ESP_022699_0985_cropped_big_spider.cub'))
# read in required data
dem.read_all()
slopes.read_all()
aspects.read_all()
spider1.read_all()
correct_azimuth(dem, aspects)
# create MarsSpicer object
mspice = MarsSpicer()
#!/Library/Frameworks/Python.framework/Versions/Current/bin/python from mars import ImgData, Point from kmaspice import MarsSpicer import os from spice import vsep,vminus import spice import numpy as np import matplotlib.pyplot as plt folder = os.environ['HOME']+'/Data/hirise/DEMs/inca' dem = ImgData(os.path.join(folder, 'big_spider_dem.cub')) # dem = ImgData(os.path.join(folder, 'DTM_Inca_City_ngate_1m_forPDS.cub')) # slopes = ImgData(os.path.join(folder,'slopes_full.tif')) # aspects = ImgData(os.path.join(folder,'aspects_full.tif')) slopes = ImgData(os.path.join(folder,'big_spider_slopes.tif')) aspects = ImgData(os.path.join(folder,'big_spider_aspects.tif')) # spider1 = ImgData(os.path.join(folder,'ESP_022607_0985_cropped_big_spider.cub')) # spider2 = ImgData(os.path.join(folder,'ESP_022699_0985_cropped_big_spider.cub')) # read in required data dem.read_all() slopes.read_all() aspects.read_all() # plt.imshow(dem.data,aspect='auto') # plt.plot(dem.center.sample,dem.center.line, 'r*',markersize=10) # determine angle between north and top to correct aspect angles that count like # north is at the top of the image newPoint = Point(lon=dem.center.lon, lat=dem.center.lat+0.001) newPoint.lonlat_to_pixel(dem.geotransform, dem.projection)
