import os
import time
start = time.time()
print('Running...')
np.set_printoptions(precision=3, suppress=True)
# Specify saving path
folder = 'dispim'
if not os.path.exists(folder):
os.makedirs(folder)
data_file = folder + '/' + folder + '.dat'
# Build microscope
exp = multi.MultiMicroscope(ill_thetas=[0, 90],
det_thetas=[90, 0],
det_nas=[0.8, 0.8],
max_l=4,
n_pts=5000)
# Calculate and save (comment this on repeat runs)
exp.calc_sys_matrix()
dill.dump(exp.psi, open(data_file, 'wb'))
# Load
f = open(data_file, 'rb')
exp.psi = dill.load(f)
# Calculate B
exp.calc_B_matrix()
# Generate phantom of true orientations
import dill
from polharmonic import util, multi, dist, sft
import os
import time
start = time.time()
print('Running...')
np.set_printoptions(precision=3, suppress=True)
# Specify saving path
folder = 'ortho'
if not os.path.exists(folder):
os.makedirs(folder)
data_file = folder + '/' + folder + '.dat'
# Build microscope
exp = multi.MultiMicroscope(ill_thetas=[90], det_nas=[1.1], max_l=4, n_pts=250)
# Calculate and save (comment this on repeat runs)
exp.calc_sys_matrix()
dill.dump(exp.psi, open(data_file, 'wb'))
import pdb
pdb.set_trace()
# Load
f = open(data_file, 'rb')
exp.psi = dill.load(f)
# Calculate B
exp.calc_B_matrix()
# Generate phantom of true orientations
import numpy as np
import dill
from polharmonic import util, multi, dist, sft
import os; import time; start = time.time(); print('Running...')
np.set_printoptions(precision=3, suppress=True)
# Specify saving path
folder = 'epi'
if not os.path.exists(folder):
os.makedirs(folder)
data_file = folder+'/'+folder+'.dat'
# Build microscope
exp = multi.MultiMicroscope(max_l=4, n_pts=5000)
# Calculate and save (comment this on repeat runs)
exp.calc_sys_matrix()
dill.dump(exp.psi, open(data_file,'wb'))
# Load
f = open(data_file, 'rb')
exp.psi = dill.load(f)
# Calculate B
exp.calc_B_matrix()
# Generate phantom of true orientations
nx, ny = 10, 10
t = np.linspace(0, np.pi/2, nx)
p = np.linspace(0, np.pi, ny)
tv, pv = np.meshgrid(t, p)
def recon_roi(name, xs, ys, zs, x0, y0, z0, recon_mask_threshold, d, skip, dpi=800, mag=1, note='', idx =[0, 3, 2, 1, 4, 5, 6, 7], file_out=None, second_cal=1):
# High level params
mfolder = 'asym_dispim'
folder = name
if not os.path.exists(folder):
os.mkdir(folder)
int_field_mag = mag
recon_mag = mag
# Import data
name_head = '/Users/Talon/GoogleDrive/projects/dispim-data/20170725_Bob_Actin_results/Cell1_LSimaging_registerred/SPIM'
names = ['A_reg_P3.tif', 'A_reg_P4.tif', 'A_reg_P1.tif', 'A_reg_P2.tif',
'B_reg_P3.tif', 'B_reg_P4.tif', 'B_reg_P1.tif', 'B_reg_P2.tif']
input_files = np.array([name_head + name for name in names])
# Calibration data
cal = np.array([1.06, 1.0, 1.0, 1.03,
1.08, 1.05, 1.0, 1.04])
# Reindexing
input_files = input_files[idx]
cal = cal[idx]
# Write to logfile
f = open(folder+'/log.txt', 'w')
f.write('name: ' + folder + '\n')
f.write('note: ' + note + '\n')
f.write('roi coords (px) = ' + str(xs) + ', ' + str(ys) + ', ' + str(zs) + '\n' )
f.write('span (px) = ' + str(xs) + ', ' + str(ys) + ', ' + str(zs) + '\n' )
f.write('span (um) = ' + str(0.135*xs) + ', ' + str(0.135*ys) + ', ' + str(0.135*zs) + '\n' )
f.write('skip (px) = ' + str(skip) + '\n')
f.write('threshold (normalized) = ' + str(recon_mask_threshold) + '\n')
f.close()
# Load and plot intensity fields
intf = data.IntensityField()
intf.load_from_file(file_names=input_files, x0=x0, y0=y0, z0=z0,
xspan=xs, yspan=ys, zspan=zs, cal=cal,
angle=0)
row_labels = ['$x$-illumination\n $z$-detection\n 1.1 NA', '$z$-illumination\n $x$-detection\n 0.71 NA']
col_labels = ['$\phi_{\mathrm{pol}} = 0^{\circ}$', '$\phi_{\mathrm{pol}} = 45^{\circ}$', '$\phi_{\mathrm{pol}} = 90^{\circ}$', '$\phi_{\mathrm{pol}} = 135^{\circ}$']
# intf.plot(output_file=folder+'/'+'data.pdf', shape=(2,4),
# row_labels=row_labels, col_labels=col_labels,
# d=d, mag=int_field_mag, dpi=400, show=False)
# Build microscopes
exp = multi.MultiMicroscope(ill_thetas=[90, 0], det_thetas=[0, 90],
det_nas=[1.1, 0.8], max_l=4, n_pts=250)
# exp_ortho = multi.MultiMicroscope(ill_thetas=[90], det_thetas=[0],
# det_nas=[1.1], max_l=4, n_pts=250)
# exp_ortho2 = multi.MultiMicroscope(ill_thetas=[0], det_thetas=[90],
# det_nas=[0.8], max_l=4, n_pts=250)
# Load asymmetric dispim system matrix
file_name = mfolder+'.dat'
data_file = mfolder+'/'+file_name
# Load system models
f = open(data_file, 'rb')
exp.psi = dill.load(f)
# exp_ortho.psi = exp.psi[:4, :]
# exp_ortho2.psi = exp.psi[4:, :]
# Calculate B
exp.calc_B_matrix()
# exp_ortho.calc_B_matrix()
# exp_ortho2.calc_B_matrix()
# Reconstruct with all data
threshold_mask = np.max(intf.g, axis=-1) > recon_mask_threshold
sparse_mask = np.zeros(threshold_mask.shape)
sparse_mask[::skip, ::skip, ::skip] = 1
mask = np.logical_and(threshold_mask, sparse_mask)
intf.g[:,:,:,4:] = intf.g[:,:,:,4:]/second_cal
df = exp.recon_dist_field(intf, mask=mask, prior='single')
df.plot_dist_field(exp.B, exp.xyz, exp.triangles,
filename=folder+'/'+file_out, r=skip, d=d,
mag=recon_mag, show=False, mask=mask)
from polharmonic import det, ill, micro, gaunt, multi
import numpy as np
m = multi.MultiMicroscope(ill_optical_axes=[[0, 0, 1]],
det_optical_axes=[[0, 0, 1]],
ill_nas=[0.8],
det_nas=[0.8],
n_samp=1.33,
ill_pols=[[1, 0, 0],
[1 / np.sqrt(2), 1 / np.sqrt(2), 0],
[0, 1, 0],
[-1 / np.sqrt(2), 1 / np.sqrt(2), 0]],
det_pols=4 * [None])
m.plot_frames(folder='pol_illum')
m = multi.MultiMicroscope(ill_optical_axes=[[0, 0, 1]],
det_optical_axes=[[0, 0, 1]],
ill_nas=[0.8],
det_nas=[0.8],
n_samp=1.33,
ill_pols=4 * [None],
det_pols=[[1, 0, 0],
[1 / np.sqrt(2), 1 / np.sqrt(2), 0],
[0, 1, 0],
[-1 / np.sqrt(2), 1 / np.sqrt(2), 0]])
m.plot_frames(folder='pol_detect')
import dill
from polharmonic import util, multi, dist, sft
import os
import time
start = time.time()
print('Running...')
np.set_printoptions(precision=3, suppress=True)
# Specify saving path
folder = 'ortho2'
if not os.path.exists(folder):
os.makedirs(folder)
data_file = folder + '/' + folder + '.dat'
# Build microscope
exp = multi.MultiMicroscope(det_thetas=[90], max_l=4, n_pts=5000)
# Calculate and save (comment this on repeat runs)
exp.calc_sys_matrix()
dill.dump(exp.psi, open(data_file, 'wb'))
# Load
f = open(data_file, 'rb')
exp.psi = dill.load(f)
# Calculate B
exp.calc_B_matrix()
# Generate phantom of true orientations
nx, ny = 10, 10
t = np.linspace(0, np.pi / 2, nx)
def recon_roi(name,
xs,
ys,
zs,
x0,
y0,
z0,
recon_mask_threshold,
d,
skip,
dpi=800,
mag=1,
note=''):
# High level params
mfolder = 'asym_dispim'
folder = name
if not os.path.exists(folder):
os.mkdir(folder)
int_field_mag = mag
recon_mag = mag
# Import data
name_head = '/Users/Talon/GoogleDrive/projects/dispim-data/20170725_Bob_Actin_results/Cell1_LSimaging_registerred/SPIM'
names = [
'A_reg_P3.tif', 'A_reg_P4.tif', 'A_reg_P1.tif', 'A_reg_P2.tif',
'B_reg_P3.tif', 'B_reg_P4.tif', 'B_reg_P1.tif', 'B_reg_P2.tif'
]
input_files = np.array([name_head + name for name in names])
# Calibration data
cal = np.array([1.06, 1.0, 1.0, 1.03, 1.08, 1.05, 1.0, 1.04])
# Reindexing
idx = [0, 3, 2, 1, 4, 5, 6, 7]
input_files = input_files[idx]
cal = cal[idx]
# Write to logfile
f = open(folder + '/log.txt', 'w')
f.write('name: ' + folder + '\n')
f.write('note: ' + note + '\n')
f.write('roi coords (px) = ' + str(xs) + ', ' + str(ys) + ', ' + str(zs) +
'\n')
f.write('span (px) = ' + str(xs) + ', ' + str(ys) + ', ' + str(zs) + '\n')
f.write('span (um) = ' + str(0.135 * xs) + ', ' + str(0.135 * ys) + ', ' +
str(0.135 * zs) + '\n')
f.write('skip (px) = ' + str(skip) + '\n')
f.write('threshold (normalized) = ' + str(recon_mask_threshold) + '\n')
f.close()
# Load and plot intensity fields
intf = data.IntensityField()
intf.load_from_file(file_names=input_files,
x0=x0,
y0=y0,
z0=z0,
xspan=xs,
yspan=ys,
zspan=zs,
cal=cal)
row_labels = [
'$x$-illumination\n $z$-detection\n 1.1 NA',
'$z$-illumination\n $x$-detection\n 0.71 NA'
]
col_labels = [
'$\phi_{\mathrm{pol}} = 0^{\circ}$',
'$\phi_{\mathrm{pol}} = 45^{\circ}$',
'$\phi_{\mathrm{pol}} = 90^{\circ}$',
'$\phi_{\mathrm{pol}} = 135^{\circ}$'
]
intf.plot(output_file=folder + '/' + 'data.pdf',
shape=(2, 4),
row_labels=row_labels,
col_labels=col_labels,
d=d,
mag=int_field_mag,
dpi=400,
show=False)
import pdb
pdb.set_trace()
# Build microscopes
exp = multi.MultiMicroscope(ill_thetas=[90, 0],
det_thetas=[0, 90],
det_nas=[1.1, 0.8],
max_l=4,
n_pts=250)
exp_ortho = multi.MultiMicroscope(ill_thetas=[90],
det_thetas=[0],
det_nas=[1.1],
max_l=4,
n_pts=250)
exp_ortho2 = multi.MultiMicroscope(ill_thetas=[0],
det_thetas=[90],
det_nas=[0.8],
max_l=4,
n_pts=250)
# Load asymmetric dispim system matrix
file_name = mfolder + '.dat'
data_file = mfolder + '/' + file_name
# Load system models
f = open(data_file, 'rb')
exp.psi = dill.load(f)
exp_ortho.psi = exp.psi[:4, :]
exp_ortho2.psi = exp.psi[4:, :]
# Calculate B
exp.calc_B_matrix()
exp_ortho.calc_B_matrix()
exp_ortho2.calc_B_matrix()
# Reconstruct with all data
threshold_mask = np.max(intf.g, axis=-1) > recon_mask_threshold
sparse_mask = np.zeros(threshold_mask.shape)
sparse_mask[::skip, ::skip, ::skip] = 1
mask = np.logical_and(threshold_mask, sparse_mask)
df = exp.recon_dist_field(intf, mask=mask, prior='single')
df.plot_dist_field(exp.B,
exp.xyz,
exp.triangles,
filename=folder + '/data_both.png',
r=skip,
d=d,
mag=recon_mag,
show=False,
mask=mask)
# Reconstruct with single view data
intf_ortho = data.IntensityField(g=intf.g[:, :, :, :4])
df_ortho = exp_ortho.recon_dist_field(intf_ortho,
mask=mask,
prior='single')
df_ortho.plot_dist_field(exp_ortho.B,
exp_ortho.xyz,
exp_ortho.triangles,
filename=folder + '/data_ortho1.png',
r=skip,
d=d,
mag=recon_mag,
show=False,
mask=mask)
intf_ortho2 = data.IntensityField(g=intf.g[:, :, :, 4:])
df_ortho2 = exp_ortho2.recon_dist_field(intf_ortho2,
mask=mask,
prior='single')
df_ortho2.plot_dist_field(exp_ortho2.B,
exp_ortho2.xyz,
exp_ortho2.triangles,
filename=folder + '/data_ortho2.png',
r=skip,
d=d,
mag=recon_mag,
show=False,
mask=mask)
# Setup viewing window
from mpl_toolkits.axes_grid.inset_locator import (inset_axes,
InsetPosition,
mark_inset)
# Plot reconstruction results
ims = [
folder + '/data_ortho1.png',
folder + '/data_ortho2.png',
folder + '/data_both.png',
]
ncols = 3
nrows = 1
inset_hfrac = .15
inset_vfrac = .15
inset_hfrac_offset = .77
inset_vfrac_offset = .32
top_pad = 0
bottom_pad = 0
left_pad = 0
right_pad = 0
hspace = .0
vspace = .0
ax_width = (1 - left_pad - right_pad - (ncols - 1) * hspace) / ncols
ax_height = (1 - top_pad - bottom_pad - (nrows - 1) * vspace) / nrows
fig = plt.figure()
ax_lst = []
for j in range(ncols):
for k in range(nrows):
a_bottom = bottom_pad + k * (ax_height + vspace)
a_left = left_pad + j * (ax_width + hspace)
inset_bottom = a_bottom + inset_vfrac_offset * ax_height
inset_left = a_left + inset_hfrac_offset * ax_width
ax = fig.add_axes([a_left, a_bottom, ax_width, ax_height])
ax_in = fig.add_axes([
inset_left, inset_bottom, ax_width * inset_hfrac,
ax_height * inset_vfrac
])
ax_lst.append((ax, ax_in))
for i, (ax, ax_in) in enumerate(ax_lst):
image = mpimg.imread(ims[i])
ax.imshow(image, interpolation=None, vmin=0, vmax=1)
ax.set_axis_off()
xyz, tp = util.fibonacci_sphere(1000, xyz=True)
sphere_im = util.plot_sphere(filename='scale.png',
directions=tp,
data=xyz,
show=False,
vis_px=500)
ax_in.imshow(sphere_im, interpolation='none')
ax_in.set_axis_off()
ax_lst[0][0].annotate('1.1 NA data only',
xy=(0, 0),
xytext=(0.5, 1.04),
textcoords='axes fraction',
va='center',
ha='center',
fontsize=7,
annotation_clip=False)
ax_lst[1][0].annotate('0.71 NA data only',
xy=(0, 0),
xytext=(0.5, 1.04),
textcoords='axes fraction',
va='center',
ha='center',
fontsize=7,
annotation_clip=False)
ax_lst[2][0].annotate('All data',
xy=(0, 0),
xytext=(0.5, 1.04),
textcoords='axes fraction',
va='center',
ha='center',
fontsize=7,
annotation_clip=False)
fig.savefig(folder + '/recon.pdf', dpi=dpi, bbox_inches='tight')
subprocess.Popen('rm ' + folder + '/*.png', shell=True)
print('Total time: ' + str(np.round(time.time() - start, 2)))
os.system('say "done"')
from polharmonic import det, ill, micro, multi
import numpy as np
#n_px=2**4 + 1
n_px = 2**4 + 1
folder = 'multi-out/'
mm = multi.MultiMicroscope(sigma_ax=0.33)
# mm.micros[0].plot(mm.micros[0].H, filename=folder+'H0.pdf', n_px=n_px, plot_m=[-2, -1, 0, 1, 2])
mm.calc_SVD(n_px=n_px)
mm.plot_SVS_3D(filename=folder + 'SVS3Dx.pdf')
mm.plot_SVS_3D(filename=folder + 'SVS3Dy.pdf',
marks=np.array([[0, 0, 0], [0, 0.5, 0], [0, 1, 0], [0, 1.5,
0]]))
mm.plot_SVS_3D(filename=folder + 'SVS3Dz.pdf',
marks=np.array([[0, 0, 0], [0, 0, 0.5], [0, 0, 1], [0, 0,
1.5]]))