def __init__(self):
self.fig = plt.figure(figsize=(3, 2))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.rgb_name_list = ['R', 'G', 'B']
self.name_not_scalable = ['r2_adjust'
] # do not apply scaler norm on those data
def __init__(self):
self.fig = plt.figure(figsize=(4,4))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.rgb_name_list = ['R', 'G', 'B']
self.ic_norm = 1e4 # multiply by this value for ic normalization
def __init__(self):
self.fig = plt.figure(figsize=(4, 4))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.rgb_name_list = ['R', 'G', 'B']
self.ic_norm = 1e4 # multiply by this value for ic normalization
def demo_rgb():
fig, ax = plt.subplots()
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
kwargs = dict(origin="lower", interpolation="nearest")
ax.imshow(im_rgb, **kwargs)
ax_r.imshow(im_r, **kwargs)
ax_g.imshow(im_g, **kwargs)
ax_b.imshow(im_b, **kwargs)
def demo_rgb():
fig, ax = plt.subplots()
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
ax.imshow(im_rgb)
ax_r.imshow(im_r)
ax_g.imshow(im_g)
ax_b.imshow(im_b)
def demo_rgb():
fig, ax = plt.subplots()
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
kwargs = dict(origin="lower", interpolation="nearest")
ax.imshow(im_rgb, **kwargs)
ax_r.imshow(im_r, **kwargs)
ax_g.imshow(im_g, **kwargs)
ax_b.imshow(im_b, **kwargs)
def demo_rgb2():
fig, ax = plt.subplots()
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
ax.imshow(im_rgb)
ax_r.imshow(im_r)
ax_g.imshow(im_g)
ax_b.imshow(im_b)
for ax in fig.axes:
ax.tick_params(axis='both', direction='in')
ax.spines[:].set_color("w")
for tick in ax.xaxis.get_major_ticks() + ax.yaxis.get_major_ticks():
tick.tick1line.set_markeredgecolor("w")
tick.tick2line.set_markeredgecolor("w")
def demo_rgb():
fig = plt.figure(1)
fig.clf()
ax = fig.add_subplot(111)
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
#fig.add_axes(ax_r)
#fig.add_axes(ax_g)
#fig.add_axes(ax_b)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
kwargs = dict(origin="lower", interpolation="nearest")
ax.imshow(im_rgb, **kwargs)
ax_r.imshow(im_r, **kwargs)
ax_g.imshow(im_g, **kwargs)
ax_b.imshow(im_b, **kwargs)
def demo_rgb():
fig = plt.figure(1)
fig.clf()
ax = fig.add_subplot(111)
ax_r, ax_g, ax_b = make_rgb_axes(ax, pad=0.02)
#fig.add_axes(ax_r)
#fig.add_axes(ax_g)
#fig.add_axes(ax_b)
r, g, b = get_rgb()
im_r, im_g, im_b, im_rgb = make_cube(r, g, b)
kwargs = dict(origin="lower", interpolation="nearest")
ax.imshow(im_rgb, **kwargs)
ax_r.imshow(im_r, **kwargs)
ax_g.imshow(im_g, **kwargs)
ax_b.imshow(im_b, **kwargs)
def show_image(self):
self.fig = plt.figure(figsize=(3, 2))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.name_not_scalable = ['r2_adjust'] # do not apply scaler norm on those data
# Check if positions data is available. Positions data may be unavailable
# (not recorded in HDF5 file) if experiment is has not been completed.
# While the data from the completed part of experiment may still be used,
# plotting vs. x-y or scatter plot may not be displayed.
positions_data_available = False
if 'positions' in self.data_dict.keys():
positions_data_available = True
# Create local copy of self.pixel_or_pos and self.grid_interpolate
pixel_or_pos_local = self.pixel_or_pos
grid_interpolate_local = self.grid_interpolate
# Disable plotting vs x-y coordinates if 'positions' data is not available
if not positions_data_available:
if pixel_or_pos_local:
pixel_or_pos_local = 0 # Switch to plotting vs. pixel number
logger.error("'Positions' data is not available. Plotting vs. x-y coordinates is disabled")
if grid_interpolate_local:
grid_interpolate_local = False # Switch to plotting vs. pixel number
logger.error("'Positions' data is not available. Interpolation is disabled.")
selected_data, selected_name = self.preprocess_data()
selected_data = np.asarray(selected_data)
if len(selected_name) != 3:
logger.error('Please select three elements for RGB plot.')
return
self.rgb_name_list = selected_name[:3]
try:
data_r = selected_data[0, :, :]
except IndexError:
selected_data = np.ones([3, 10, 10])
def _compute_equal_axes_ranges(x_min, x_max, y_min, y_max):
"""
Compute ranges for x- and y- axes of the plot. Make sure that the ranges for x- and y-axes are
always equal and fit the maximum of the ranges for x and y values:
max(abs(x_max-x_min), abs(y_max-y_min))
The ranges are set so that the data is always centered in the middle of the ranges
Parameters
----------
x_min, x_max, y_min, y_max : float
lower and upper boundaries of the x and y values
Returns
-------
x_axis_min, x_axis_max, y_axis_min, y_axis_max : float
lower and upper boundaries of the x- and y-axes ranges
"""
x_axis_min, x_axis_max, y_axis_min, y_axis_max = x_min, x_max, y_min, y_max
x_range, y_range = abs(x_max - x_min), abs(y_max - y_min)
if x_range > y_range:
y_center = (y_max + y_min) / 2
y_axis_max = y_center + x_range / 2
y_axis_min = y_center - x_range / 2
else:
x_center = (x_max + x_min) / 2
x_axis_max = x_center + y_range / 2
x_axis_min = x_center - y_range / 2
return x_axis_min, x_axis_max, y_axis_min, y_axis_max
def _adjust_data_range_using_min_ratio(c_min, c_max, c_axis_range, *, min_ratio=0.01):
"""
Adjust the range for plotted data along one axis (x or y). The adjusted range is
applied to the 'extent' attribute of imshow(). The adjusted range is always greater
than 'axis_range * min_ratio'. Such transformation has no physical meaning
and performed for aesthetic reasons: stretching the image presentation of
a scan with only a few lines (1-3) greatly improves visibility of data.
Parameters
----------
c_min, c_max : float
boundaries of the data range (along x or y axis)
c_axis_range : float
range presented along the same axis
Returns
-------
cmin, c_max : float
adjusted boundaries of the data range
"""
c_range = c_max - c_min
if c_range < c_axis_range * min_ratio:
c_center = (c_max + c_min) / 2
c_new_range = c_axis_range * min_ratio
c_min = c_center - c_new_range / 2
c_max = c_center + c_new_range / 2
return c_min, c_max
if pixel_or_pos_local:
# xd_min, xd_max, yd_min, yd_max = min(self.x_pos), max(self.x_pos),
# min(self.y_pos), max(self.y_pos)
x_pos_2D = self.data_dict['positions']['x_pos']
y_pos_2D = self.data_dict['positions']['y_pos']
xd_min, xd_max, yd_min, yd_max = x_pos_2D.min(), x_pos_2D.max(), y_pos_2D.min(), y_pos_2D.max()
xd_axis_min, xd_axis_max, yd_axis_min, yd_axis_max = \
_compute_equal_axes_ranges(xd_min, xd_max, yd_min, yd_max)
xd_min, xd_max = _adjust_data_range_using_min_ratio(xd_min, xd_max, xd_axis_max - xd_axis_min)
yd_min, yd_max = _adjust_data_range_using_min_ratio(yd_min, yd_max, yd_axis_max - yd_axis_min)
# Adjust the direction of each axis depending on the direction in which encoder values changed
# during the experiment. Data is plotted starting from the upper-right corner of the plot
if x_pos_2D[0, 0] > x_pos_2D[0, -1]:
xd_min, xd_max, xd_axis_min, xd_axis_max = xd_max, xd_min, xd_axis_max, xd_axis_min
if y_pos_2D[0, 0] > y_pos_2D[-1, 0]:
yd_min, yd_max, yd_axis_min, yd_axis_max = yd_max, yd_min, yd_axis_max, yd_axis_min
else:
# Set equal ranges for the axes data
yd, xd = selected_data.shape[1], selected_data.shape[2]
xd_min, xd_max, yd_min, yd_max = 0, xd, 0, yd
# Select minimum range for data
if (yd <= math.floor(xd / 100)) and (xd >= 200):
yd_min, yd_max = -math.floor(xd / 200), math.ceil(xd / 200)
if (xd <= math.floor(yd / 100)) and (yd >= 200):
xd_min, xd_max = -math.floor(yd / 200), math.ceil(yd / 200)
xd_axis_min, xd_axis_max, yd_axis_min, yd_axis_max = \
_compute_equal_axes_ranges(xd_min, xd_max, yd_min, yd_max)
name_r = self.rgb_name_list[self.index_red]
data_r = selected_data[self.index_red, :, :]
name_g = self.rgb_name_list[self.index_green]
data_g = selected_data[self.index_green, :, :]
name_b = self.rgb_name_list[self.index_blue]
data_b = selected_data[self.index_blue, :, :]
rgb_l_h = ({'low': self.r_low, 'high': self.r_high},
{'low': self.g_low, 'high': self.g_high},
{'low': self.b_low, 'high': self.b_high})
def _norm_data(data):
"""
Normalize data between (0, 1).
Parameters
----------
data : 2D array
"""
data_min = np.min(data)
c_norm = np.max(data) - data_min
return (data - data_min) / c_norm if (c_norm != 0) else (data - data_min)
def _stretch_range(data_in, v_low, v_high):
# 'data is already normalized, so that the values are in the range 0..1
# v_low, v_high are in the range 0..100
if (v_low <= 0) and (v_high >= 100):
return data_in
if v_high - v_low < 1: # This should not happen in practice, but check just in case
v_high = v_low + 1
v_low, v_high = v_low / 100.0, v_high / 100.0
c = 1.0 / (v_high - v_low)
data_out = (data_in - v_low) * c
return np.clip(data_out, 0, 1.0)
# Interpolate non-uniformly spaced data to uniform grid
if grid_interpolate_local:
data_r, _, _ = grid_interpolate(data_r,
self.data_dict['positions']['x_pos'],
self.data_dict['positions']['y_pos'])
data_g, _, _ = grid_interpolate(data_g,
self.data_dict['positions']['x_pos'],
self.data_dict['positions']['y_pos'])
data_b, _, _ = grid_interpolate(data_b,
self.data_dict['positions']['x_pos'],
self.data_dict['positions']['y_pos'])
# Normalize data
data_r = _norm_data(data_r)
data_g = _norm_data(data_g)
data_b = _norm_data(data_b)
data_r = _stretch_range(data_r, rgb_l_h[self.index_red]['low'], rgb_l_h[self.index_red]['high'])
data_g = _stretch_range(data_g, rgb_l_h[self.index_green]['low'], rgb_l_h[self.index_green]['high'])
data_b = _stretch_range(data_b, rgb_l_h[self.index_blue]['low'], rgb_l_h[self.index_blue]['high'])
R, G, B, RGB = make_cube(data_r,
data_g,
data_b)
red_patch = mpatches.Patch(color='red', label=name_r)
green_patch = mpatches.Patch(color='green', label=name_g)
blue_patch = mpatches.Patch(color='blue', label=name_b)
kwargs = dict(origin="upper", interpolation="nearest", extent=(xd_min, xd_max, yd_max, yd_min))
self.ax.imshow(RGB, **kwargs)
self.ax_r.imshow(R, **kwargs)
self.ax_g.imshow(G, **kwargs)
self.ax_b.imshow(B, **kwargs)
self.ax.set_xlim(xd_axis_min, xd_axis_max)
self.ax.set_ylim(yd_axis_max, yd_axis_min)
self.ax_r.set_xlim(xd_axis_min, xd_axis_max)
self.ax_r.set_ylim(yd_axis_max, yd_axis_min)
self.ax_g.set_xlim(xd_axis_min, xd_axis_max)
self.ax_g.set_ylim(yd_axis_max, yd_axis_min)
self.ax_b.set_xlim(xd_axis_min, xd_axis_max)
self.ax_b.set_ylim(yd_axis_max, yd_axis_min)
self.ax.xaxis.set_major_locator(mticker.MaxNLocator(nbins="auto"))
self.ax.yaxis.set_major_locator(mticker.MaxNLocator(nbins="auto"))
plt.setp(self.ax_r.get_xticklabels(), visible=False)
plt.setp(self.ax_r.get_yticklabels(), visible=False)
plt.setp(self.ax_g.get_xticklabels(), visible=False)
plt.setp(self.ax_g.get_yticklabels(), visible=False)
plt.setp(self.ax_b.get_xticklabels(), visible=False)
plt.setp(self.ax_b.get_yticklabels(), visible=False)
# self.ax_r.set_xticklabels([])
# self.ax_r.set_yticklabels([])
# sb_x = 38
# sb_y = 46
# sb_length = 10
# sb_height = 1
# ax.add_patch(mpatches.Rectangle(( sb_x, sb_y), sb_length, sb_height, color='white'))
# ax.text(sb_x + sb_length /2, sb_y - 1*sb_height, '100 nm', color='w', ha='center',
# va='bottom', backgroundcolor='black', fontsize=18)
self.ax.legend(bbox_to_anchor=(0., 1.0, 1., .10), ncol=3,
handles=[red_patch, green_patch, blue_patch], mode="expand", loc=3)
# self.fig.tight_layout(pad=4.0, w_pad=0.8, h_pad=0.8)
# self.fig.tight_layout()
# self.fig.canvas.draw_idle()
self.fig.suptitle(self.img_title, fontsize=20)
self.fig.canvas.draw_idle()
def __init__(self):
self.fig = plt.figure(figsize=(4, 4))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.rgb_name_list = ['R', 'G', 'B']
def __init__(self):
self.fig = plt.figure(figsize=(3,2))
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
self.rgb_name_list = ['R', 'G', 'B']
self.name_not_scalable = ['r2_adjust'] # do not apply scaler norm on those data
def show_image(self):
# Don't plot the image if dictionary is empty (causes a lot of issues)
if not self.io_model.img_dict:
return
self.fig.clf()
self.ax = self.fig.add_subplot(111)
self.ax_r, self.ax_g, self.ax_b = make_rgb_axes(self.ax, pad=0.02)
# Check if positions data is available. Positions data may be unavailable
# (not recorded in HDF5 file) if experiment is has not been completed.
# While the data from the completed part of experiment may still be used,
# plotting vs. x-y or scatter plot may not be displayed.
positions_data_available = False
if "positions" in self.io_model.img_dict.keys():
positions_data_available = True
# Create local copy of self.pixel_or_pos and self.grid_interpolate
pixel_or_pos_local = self.pixel_or_pos
grid_interpolate_local = self.grid_interpolate
# Disable plotting vs x-y coordinates if 'positions' data is not available
if not positions_data_available:
if pixel_or_pos_local:
pixel_or_pos_local = 0 # Switch to plotting vs. pixel number
logger.error(
"'Positions' data is not available. Plotting vs. x-y coordinates is disabled"
)
if grid_interpolate_local:
grid_interpolate_local = False # Switch to plotting vs. pixel number
logger.error(
"'Positions' data is not available. Interpolation is disabled."
)
selected_data, selected_names, rgb_color_to_keys, quant_norm_applied = self.preprocess_data(
)
selected_data = np.asarray(selected_data)
# Hide unused axes
if rgb_color_to_keys["red"] is None:
self.ax_r.set_visible(False)
if rgb_color_to_keys["green"] is None:
self.ax_g.set_visible(False)
if rgb_color_to_keys["blue"] is None:
self.ax_b.set_visible(False)
if selected_data.ndim != 3:
# There is no data to display. Hide the last axis and exit
self.ax.set_visible(False)
return
def _compute_equal_axes_ranges(x_min, x_max, y_min, y_max):
"""
Compute ranges for x- and y- axes of the plot. Make sure that the ranges for x- and y-axes are
always equal and fit the maximum of the ranges for x and y values:
max(abs(x_max-x_min), abs(y_max-y_min))
The ranges are set so that the data is always centered in the middle of the ranges
Parameters
----------
x_min, x_max, y_min, y_max : float
lower and upper boundaries of the x and y values
Returns
-------
x_axis_min, x_axis_max, y_axis_min, y_axis_max : float
lower and upper boundaries of the x- and y-axes ranges
"""
x_axis_min, x_axis_max, y_axis_min, y_axis_max = x_min, x_max, y_min, y_max
x_range, y_range = abs(x_max - x_min), abs(y_max - y_min)
if x_range > y_range:
y_center = (y_max + y_min) / 2
y_axis_max = y_center + x_range / 2
y_axis_min = y_center - x_range / 2
else:
x_center = (x_max + x_min) / 2
x_axis_max = x_center + y_range / 2
x_axis_min = x_center - y_range / 2
return x_axis_min, x_axis_max, y_axis_min, y_axis_max
def _adjust_data_range_using_min_ratio(c_min,
c_max,
c_axis_range,
*,
min_ratio=0.01):
"""
Adjust the range for plotted data along one axis (x or y). The adjusted range is
applied to the 'extent' attribute of imshow(). The adjusted range is always greater
than 'axis_range * min_ratio'. Such transformation has no physical meaning
and performed for aesthetic reasons: stretching the image presentation of
a scan with only a few lines (1-3) greatly improves visibility of data.
Parameters
----------
c_min, c_max : float
boundaries of the data range (along x or y axis)
c_axis_range : float
range presented along the same axis
Returns
-------
cmin, c_max : float
adjusted boundaries of the data range
"""
c_range = c_max - c_min
if c_range < c_axis_range * min_ratio:
c_center = (c_max + c_min) / 2
c_new_range = c_axis_range * min_ratio
c_min = c_center - c_new_range / 2
c_max = c_center + c_new_range / 2
return c_min, c_max
if pixel_or_pos_local:
# xd_min, xd_max, yd_min, yd_max = min(self.x_pos), max(self.x_pos),
# min(self.y_pos), max(self.y_pos)
x_pos_2D = self.io_model.img_dict["positions"]["x_pos"]
y_pos_2D = self.io_model.img_dict["positions"]["y_pos"]
xd_min, xd_max, yd_min, yd_max = x_pos_2D.min(), x_pos_2D.max(
), y_pos_2D.min(), y_pos_2D.max()
xd_axis_min, xd_axis_max, yd_axis_min, yd_axis_max = _compute_equal_axes_ranges(
xd_min, xd_max, yd_min, yd_max)
xd_min, xd_max = _adjust_data_range_using_min_ratio(
xd_min, xd_max, xd_axis_max - xd_axis_min)
yd_min, yd_max = _adjust_data_range_using_min_ratio(
yd_min, yd_max, yd_axis_max - yd_axis_min)
# Adjust the direction of each axis depending on the direction in which encoder values changed
# during the experiment. Data is plotted starting from the upper-right corner of the plot
if x_pos_2D[0, 0] > x_pos_2D[0, -1]:
xd_min, xd_max, xd_axis_min, xd_axis_max = xd_max, xd_min, xd_axis_max, xd_axis_min
if y_pos_2D[0, 0] > y_pos_2D[-1, 0]:
yd_min, yd_max, yd_axis_min, yd_axis_max = yd_max, yd_min, yd_axis_max, yd_axis_min
else:
if selected_data.ndim == 3:
# Set equal ranges for the axes data
yd, xd = selected_data.shape[1], selected_data.shape[2]
xd_min, xd_max, yd_min, yd_max = 0, xd, 0, yd
# Select minimum range for data
if (yd <= math.floor(xd / 100)) and (xd >= 200):
yd_min, yd_max = -math.floor(xd / 200), math.ceil(xd / 200)
if (xd <= math.floor(yd / 100)) and (yd >= 200):
xd_min, xd_max = -math.floor(yd / 200), math.ceil(yd / 200)
xd_axis_min, xd_axis_max, yd_axis_min, yd_axis_max = _compute_equal_axes_ranges(
xd_min, xd_max, yd_min, yd_max)
name_r, data_r, limits_r = "", None, {"low": 0, "high": 100.0}
name_g, data_g, limits_g = "", None, {"low": 0, "high": 100.0}
name_b, data_b, limits_b = "", None, {"low": 0, "high": 100.0}
for color, name in rgb_color_to_keys.items():
if name:
try:
ind = selected_names.index(name)
name_label = name
if quant_norm_applied[ind]:
name_label += " - Q" # Add suffix to name if quantitative normalization was applied
if color == "red":
name_r, data_r = name_label, selected_data[ind]
limits_r = self.limit_dict[name]
elif color == "green":
name_g, data_g = name_label, selected_data[ind]
limits_g = self.limit_dict[name]
elif color == "blue":
name_b, data_b = name_label, selected_data[ind]
limits_b = self.limit_dict[name]
except ValueError:
pass
def _norm_data(data):
"""
Normalize data between (0, 1).
Parameters
----------
data : 2D array
"""
if data is None:
return data
data_min = np.min(data)
c_norm = np.max(data) - data_min
return (data - data_min) / c_norm if (c_norm != 0) else (data -
data_min)
def _stretch_range(data_in, v_low, v_high):
# 'data is already normalized, so that the values are in the range 0..1
# v_low, v_high are in the range 0..100
if data_in is None:
return data_in
if (v_low <= 0) and (v_high >= 100):
return data_in
if v_high - v_low < 1: # This should not happen in practice, but check just in case
v_high = v_low + 1
v_low, v_high = v_low / 100.0, v_high / 100.0
c = 1.0 / (v_high - v_low)
data_out = (data_in - v_low) * c
return np.clip(data_out, 0, 1.0)
# Interpolate non-uniformly spaced data to uniform grid
if grid_interpolate_local:
data_r, _, _ = grid_interpolate(
data_r, self.io_model.img_dict["positions"]["x_pos"],
self.io_model.img_dict["positions"]["y_pos"])
data_g, _, _ = grid_interpolate(
data_g, self.io_model.img_dict["positions"]["x_pos"],
self.io_model.img_dict["positions"]["y_pos"])
data_b, _, _ = grid_interpolate(
data_b, self.io_model.img_dict["positions"]["x_pos"],
self.io_model.img_dict["positions"]["y_pos"])
# The dictionaries 'rgb_view_data' and 'pos_limits' are used for monitoring
# the map values at current cursor positions.
rgb_view_data = {_: None for _ in self.rgb_keys}
if data_r is not None:
rgb_view_data["red"] = data_r
if data_g is not None:
rgb_view_data["green"] = data_g
if data_b is not None:
rgb_view_data["blue"] = data_b
pos_limits = {
"x_low": xd_min,
"x_high": xd_max,
"y_low": yd_min,
"y_high": yd_max
}
# Normalize data
data_r_norm = _norm_data(data_r)
data_g_norm = _norm_data(data_g)
data_b_norm = _norm_data(data_b)
data_r_norm = _stretch_range(data_r_norm, limits_r["low"],
limits_r["high"])
data_g_norm = _stretch_range(data_g_norm, limits_g["low"],
limits_g["high"])
data_b_norm = _stretch_range(data_b_norm, limits_b["low"],
limits_b["high"])
R, G, B, RGB = make_cube(data_r_norm, data_g_norm, data_b_norm)
red_patch = mpatches.Patch(color="red", label=name_r)
green_patch = mpatches.Patch(color="green", label=name_g)
blue_patch = mpatches.Patch(color="blue", label=name_b)
def format_coord_func(x,
y,
*,
pixel_or_pos,
rgb_color_to_keys,
rgb_view_data,
pos_limits,
colors=None):
x0, y0 = pos_limits["x_low"], pos_limits["y_low"]
if colors is None:
colors = list(rgb_color_to_keys.keys())
s = ""
for n, color in enumerate(self.rgb_keys):
if (color
not in colors) or (rgb_color_to_keys[color] is None
) or (rgb_view_data[color] is None):
continue
map = rgb_view_data[color]
ny, nx = map.shape
dy = (pos_limits["y_high"] - y0) / ny if ny else 0
dx = (pos_limits["x_high"] - x0) / nx if nx else 0
cy = 1 / dy if dy else 1
cx = 1 / dx if dx else 1
x_pixel = math.floor((x - x0) * cx)
y_pixel = math.floor((y - y0) * cy)
if (0 <= x_pixel < nx) and (0 <= y_pixel < ny):
# The following line is extremely useful for debugging the feature. Keep it.
# s += f" <b>{rgb_color_to_keys[color]}</b>: {x_pixel} {y_pixel}"
s += f" <b>{rgb_color_to_keys[color]}</b>: {map[y_pixel, x_pixel]:.5g}"
s = " - " + s if s else s # Add dash if something is to be printed
if pixel_or_pos:
# Spatial coordinates (double)
s_coord = f"({x:.5g}, {y:.5g})"
else:
# Pixel coordinates (int)
s_coord = f"({int(x)}, {int(y)})"
return s_coord + s
format_coord = partial(
format_coord_func,
pixel_or_pos=pixel_or_pos_local,
rgb_color_to_keys=rgb_color_to_keys,
rgb_view_data=rgb_view_data,
pos_limits=pos_limits,
)
def format_cursor_data(data):
return "" # Print nothing
kwargs = dict(origin="upper",
interpolation="nearest",
extent=(xd_min, xd_max, yd_max, yd_min))
if RGB is not None:
img = self.ax.imshow(RGB, **kwargs)
self.ax.format_coord = format_coord
img.format_cursor_data = format_cursor_data
self.ax.set_xlim(xd_axis_min, xd_axis_max)
self.ax.set_ylim(yd_axis_max, yd_axis_min)
if R is not None:
img = self.ax_r.imshow(R, **kwargs)
self.ax_r.set_xlim(xd_axis_min, xd_axis_max)
self.ax_r.set_ylim(yd_axis_max, yd_axis_min)
format_coord_r = partial(format_coord, colors=["red"])
self.ax_r.format_coord = format_coord_r
img.format_cursor_data = format_cursor_data
if G is not None:
img = self.ax_g.imshow(G, **kwargs)
self.ax_g.set_xlim(xd_axis_min, xd_axis_max)
self.ax_g.set_ylim(yd_axis_max, yd_axis_min)
format_coord_g = partial(format_coord, colors=["green"])
self.ax_g.format_coord = format_coord_g
img.format_cursor_data = format_cursor_data
if B is not None:
img = self.ax_b.imshow(B, **kwargs)
self.ax_b.set_xlim(xd_axis_min, xd_axis_max)
self.ax_b.set_ylim(yd_axis_max, yd_axis_min)
format_coord_b = partial(format_coord, colors=["blue"])
self.ax_b.format_coord = format_coord_b
img.format_cursor_data = format_cursor_data
self.ax.xaxis.set_major_locator(mticker.MaxNLocator(nbins="auto"))
self.ax.yaxis.set_major_locator(mticker.MaxNLocator(nbins="auto"))
plt.setp(self.ax_r.get_xticklabels(), visible=False)
plt.setp(self.ax_r.get_yticklabels(), visible=False)
plt.setp(self.ax_g.get_xticklabels(), visible=False)
plt.setp(self.ax_g.get_yticklabels(), visible=False)
plt.setp(self.ax_b.get_xticklabels(), visible=False)
plt.setp(self.ax_b.get_yticklabels(), visible=False)
# self.ax_r.set_xticklabels([])
# self.ax_r.set_yticklabels([])
# sb_x = 38
# sb_y = 46
# sb_length = 10
# sb_height = 1
# ax.add_patch(mpatches.Rectangle(( sb_x, sb_y), sb_length, sb_height, color='white'))
# ax.text(sb_x + sb_length /2, sb_y - 1*sb_height, '100 nm', color='w', ha='center',
# va='bottom', backgroundcolor='black', fontsize=18)
self.ax_r.legend(
loc="upper left",
bbox_to_anchor=(1.1, 0),
frameon=False,
handles=[red_patch, green_patch, blue_patch],
mode="expand",
)
# self.fig.tight_layout(pad=4.0, w_pad=0.8, h_pad=0.8)
# self.fig.tight_layout()
# self.fig.canvas.draw_idle()
# self.fig.suptitle(self.img_title, fontsize=20)
self.fig.canvas.draw_idle()
def plot2d_rgbgrid(self,
axlim=25,
interp_method='lanczos',
pix_grid=None,
fig=None,
ax=None):
"""Create a 2D color plot of the PSF and R,G,B components.
Parameters
----------
axlim : `float`
limits of axis, symmetric. xlim=(-axlim,axlim), ylim=(-axlim, axlim)
interp_method : `str`
method used to interpolate the image between samples of the PSF
pix_grid : float
if not None, overlays gridlines with spacing equal to pix_grid.
Intended to show the collection into camera pixels while still in
the oversampled domain
fig : `matplotlib.figure.Figure`, optional
Figure containing the plot
ax : `matplotlib.axes.Axis`, optional:
Axis containing the plot
fig : `matplotlib.figure.Figure`, optional
Figure containing the plot
ax : `matplotlib.axes.Axis`, optional:
Axis containing the plot
Notes
-----
Need to refine internal workings at some point.
"""
# make the arrays for the RGB images
dat = m.empty((self.samples_y, self.samples_x, 3))
datr = m.zeros((self.samples_y, self.samples_x, 3))
datg = m.zeros((self.samples_y, self.samples_x, 3))
datb = m.zeros((self.samples_y, self.samples_x, 3))
dat[:, :, 0] = self.R
dat[:, :, 1] = self.G
dat[:, :, 2] = self.B
datr[:, :, 0] = self.R
datg[:, :, 1] = self.G
datb[:, :, 2] = self.B
left, right = self.unit[0], self.unit[-1]
ax_width = 2 * axlim
# generate a figure and axes to plot in
fig, ax = share_fig_ax(fig, ax)
axr, axg, axb = make_rgb_axes(ax)
ax.imshow(dat,
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axr.imshow(datr,
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axg.imshow(datg,
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axb.imshow(datb,
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
for axs in (ax, axr, axg, axb):
ax.set(xlim=(-axlim, axlim), ylim=(-axlim, axlim))
if pix_grid is not None:
# if pixel grid is desired, add it
mult = m.m.floor(axlim / pix_grid)
gmin, gmax = -mult * pix_grid, mult * pix_grid
pts = m.arange(gmin, gmax, pix_grid)
ax.set_yticks(pts, minor=True)
ax.set_xticks(pts, minor=True)
ax.yaxis.grid(True, which='minor')
ax.xaxis.grid(True, which='minor')
ax.set(xlabel=r'Image Plane X [$\mu m$]',
ylabel=r'Image Plane Y [$\mu m$]')
axr.text(-axlim + 0.1 * ax_width,
axlim - 0.2 * ax_width,
'R',
color='white')
axg.text(-axlim + 0.1 * ax_width,
axlim - 0.2 * ax_width,
'G',
color='white')
axb.text(-axlim + 0.1 * ax_width,
axlim - 0.2 * ax_width,
'B',
color='white')
return fig, ax
def plot2d_rgbgrid(self, axlim=25, interp_method='lanczos',
pix_grid=None, fig=None, ax=None):
'''Creates a 2D color plot of the PSF and components
Args:
axlim (`float`): limits of axis, symmetric.
xlim=(-axlim,axlim), ylim=(-axlim, axlim).
interp_method (string): method used to interpolate the image between
samples of the PSF.
pix_grid (float): if not None, overlays gridlines with spacing equal
to pix_grid. Intended to show the collection into camera pixels
while still in the oversampled domain.
fig (pyplot.figure): figure to plot in.
ax (pyplot.axis): axis to plot in.
Returns:
pyplot.fig, pyplot.axis. Figure and axis containing the plot.
Notes:
Need to refine internal workings at some point.
'''
# make the arrays for the RGB images
dat = np.empty((self.samples_y, self.samples_x, 3))
datr = np.zeros((self.samples_y, self.samples_x, 3))
datg = np.zeros((self.samples_y, self.samples_x, 3))
datb = np.zeros((self.samples_y, self.samples_x, 3))
dat[:, :, 0] = self.R
dat[:, :, 1] = self.G
dat[:, :, 2] = self.B
datr[:, :, 0] = self.R
datg[:, :, 1] = self.G
datb[:, :, 2] = self.B
left, right = self.unit[0], self.unit[-1]
ax_width = 2 * axlim
# generate a figure and axes to plot in
fig, ax = share_fig_ax(fig, ax)
axr, axg, axb = make_rgb_axes(ax)
ax.imshow(correct_gamma(dat),
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axr.imshow(correct_gamma(datr),
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axg.imshow(correct_gamma(datg),
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
axb.imshow(correct_gamma(datb),
extent=[left, right, left, right],
interpolation=interp_method,
origin='lower')
for axs in (ax, axr, axg, axb):
ax.set(xlim=(-axlim, axlim), ylim=(-axlim, axlim))
if pix_grid is not None:
# if pixel grid is desired, add it
mult = np.floor(axlim / pix_grid)
gmin, gmax = -mult * pix_grid, mult * pix_grid
pts = np.arange(gmin, gmax, pix_grid)
ax.set_yticks(pts, minor=True)
ax.set_xticks(pts, minor=True)
ax.yaxis.grid(True, which='minor')
ax.xaxis.grid(True, which='minor')
ax.set(xlabel=r'Image Plane X [$\mu m$]', ylabel=r'Image Plane Y [$\mu m$]')
axr.text(-axlim + 0.1 * ax_width, axlim - 0.2 * ax_width, 'R', color='white')
axg.text(-axlim + 0.1 * ax_width, axlim - 0.2 * ax_width, 'G', color='white')
axb.text(-axlim + 0.1 * ax_width, axlim - 0.2 * ax_width, 'B', color='white')
return fig, ax
