def getDistance(self, pixels1, pixels2):
lab1 = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(pixels1))
lab2 = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(pixels2))
dist = mean_squared_error(pixels1, pixels2)
#dist = np.sqrt((pixels1[:, 0] - pixels2[:, 0])**2 + (pixels1[:, 1] - pixels2[:, 1])**2 + (pixels1[:, 2] - pixels2[:, 2])**2)
#dist = np.sum(dist)
return dist
def update_CIELAB_3Dplot(fig, tristimulus_XYZ, selected_phi, data):
if fig is None or tristimulus_XYZ is None or selected_phi is None:
raise PreventUpdate
selected_phi = np.array(selected_phi)
phiVs = np.array(data['phiVs'])
tristimulus_XYZ = np.array(tristimulus_XYZ)
figure = go.Figure()
if selected_phi.shape[0] == 1:
selected_XYZ = tristimulus_XYZ[:, phiVs == selected_phi[0]][:, 0, :]
selected_LAB = np.array([
clr.XYZ_to_Lab(selected_XYZ[i] / 100)
for i in range(selected_XYZ.shape[0])
])
figure.add_trace(
go.Scatter3d(x=selected_LAB[:, 1],
y=selected_LAB[:, 2],
z=selected_LAB[:, 0],
mode='lines+markers'))
elif selected_phi.shape[0] == 2:
selected_XYZ_pos = tristimulus_XYZ[:, phiVs == selected_phi[0]][:,
0, :]
selected_LAB_pos = np.array([
clr.XYZ_to_Lab(selected_XYZ_pos[i] / 100)
for i in range(selected_XYZ_pos.shape[0])
])
figure.add_trace(
go.Scatter3d(x=selected_LAB_pos[:, 1],
y=selected_LAB_pos[:, 2],
z=selected_LAB_pos[:, 0],
mode='lines+markers'))
selected_XYZ_neg = tristimulus_XYZ[:, phiVs == selected_phi[1]][:,
0, :]
selected_XYZ_neg = np.array([
clr.XYZ_to_Lab(selected_XYZ_neg[i] / 100)
for i in range(selected_XYZ_neg.shape[0])
])
figure.add_trace(
go.Scatter3d(x=selected_XYZ_neg[:, 1],
y=selected_XYZ_neg[:, 2],
z=selected_XYZ_neg[:, 0],
mode='lines+markers'))
else:
raise PreventUpdate
figure.update_layout(title="Color travel 3D plot in CIELab space",
scene=dict(xaxis_title="a*",
yaxis_title="b*",
zaxis_title="L*"))
return figure
def update_CIELAB_3Dplot(fig, tristimulus_XYZ, selected_phi, data):
if fig is None or tristimulus_XYZ is None or selected_phi is None:
raise PreventUpdate
selected_phi = np.array(selected_phi)
phiVs = np.array(data['phiVs'])
tristimulus_XYZ = np.array(tristimulus_XYZ)
figure = go.Figure()
if selected_phi.shape[0] == 1:
selected_XYZ = tristimulus_XYZ[:, phiVs == selected_phi[0]][:, 0, :]
selected_LAB = np.array([
clr.XYZ_to_Lab(selected_XYZ[i] / 100)
for i in range(selected_XYZ.shape[0])
])
figure.add_trace(
go.Scatter(name='projection',
x=selected_LAB[:, 1],
y=selected_LAB[:, 2],
mode='lines+markers'))
elif selected_phi.shape[0] == 2:
selected_XYZ_pos = tristimulus_XYZ[:, phiVs == selected_phi[0]][:,
0, :]
selected_LAB_pos = np.array([
clr.XYZ_to_Lab(selected_XYZ_pos[i] / 100)
for i in range(selected_XYZ_pos.shape[0])
])
figure.add_trace(
go.Scatter(name='projection 0 to 90',
x=selected_LAB_pos[:, 1],
y=selected_LAB_pos[:, 2],
mode='lines+markers'))
selected_XYZ_neg = tristimulus_XYZ[:, phiVs == selected_phi[1]][:,
0, :]
selected_XYZ_neg = np.array([
clr.XYZ_to_Lab(selected_XYZ_neg[i] / 100)
for i in range(selected_XYZ_neg.shape[0])
])
figure.add_trace(
go.Scatter(name='projection -90 to 0',
x=selected_XYZ_neg[:, 1],
y=selected_XYZ_neg[:, 2],
mode='lines+markers'))
else:
raise PreventUpdate
figure.update_layout(title="CIELab colorspace projection to a*b* plane",
xaxis_title='a*',
yaxis_title='b*')
return figure
def delta_E_range():
XYZ = colour.volume.XYZ_outer_surface()
# print(XYZ, XYZ.shape)
combinations = colour.XYZ_to_Lab(
np.array(list(itertools.combinations(XYZ, 2))))
# print(combinations)
delta_E = colour.delta_E(combinations[:, 0, :], combinations[:, 1, :])
return np.amax(delta_E), np.amin(delta_E)
def convert(dict, illuminant='D65'):
""" Converts the set of XYZ colours into other spaces: Lab, LCH, sRGB """
illuminant = colour.ILLUMINANTS['CIE 1931 2 Degree Standard Observer'][
illuminant]
dict['Lab'] = colour.XYZ_to_Lab(dict['XYZ'] / 100.,
illuminant) # beware: expects XYZ in [0,1]
dict['LCH'] = colour.Lab_to_LCHab(dict['Lab'])
dict['RGB'] = colour.XYZ_to_sRGB(
dict['XYZ'] / 100., illuminant) # beware: expects XYZ in [0,1]
def calculateColorValues(splineT, splineR, settings):
'''Function calculates color values of the Transmission and Refelection side of the stack.
Input Arguments are tranmission and reflection spline functions.
Returns array of values/tuples of different standards.'''
import colour
import numpy as np
wvl = np.linspace(380, 780, 81)
dic_T = {}
dic_R = {}
dic_test = {}
for idx, value in enumerate(wvl):
dic_T[value] = splineT(value).item()
dic_R[value] = splineR(value).item()
#Removes warnings from conversions.
colour.filter_warnings()
cmfs = colour.STANDARD_OBSERVERS_CMFS[settings.color_cmfs] #1931 etc
illuminant = colour.ILLUMINANTS_RELATIVE_SPDS[
settings.color_illuminant] #D65, A, C
T_spd = colour.SpectralPowerDistribution('', dic_T)
T_XYZ = colour.spectral_to_XYZ(T_spd, cmfs, illuminant)
T_xy = colour.XYZ_to_xy(T_XYZ / 100)
T_ab = colour.XYZ_to_Lab(T_XYZ / 100,
illuminant=colour.ILLUMINANTS[settings.color_cmfs]
[settings.color_illuminant])
T_rgb = colour.XYZ_to_sRGB(
T_XYZ / 100,
illuminant=colour.ILLUMINANTS[settings.color_cmfs][
settings.color_illuminant])
R_spd = colour.SpectralPowerDistribution('', dic_R)
R_XYZ = colour.spectral_to_XYZ(R_spd, cmfs, illuminant)
R_xy = colour.XYZ_to_xy(R_XYZ / 100)
R_ab = colour.XYZ_to_Lab(R_XYZ / 100,
illuminant=colour.ILLUMINANTS[settings.color_cmfs]
[settings.color_illuminant])
R_rgb = colour.XYZ_to_sRGB(
R_XYZ / 100,
illuminant=colour.ILLUMINANTS[settings.color_cmfs][
settings.color_illuminant])
return (T_XYZ, T_xy, T_ab, T_rgb, R_XYZ, R_xy, R_ab, R_rgb)
def rgb_to_lab_d65(rgb, name="ITU-R BT.709"):
illuminant_XYZ = tpg.D65_WHITE
illuminant_RGB = tpg.D65_WHITE
chromatic_adaptation_transform = 'CAT02'
rgb_to_xyz_matrix = tpg.get_rgb_to_xyz_matrix(name)
large_xyz = colour.RGB_to_XYZ(rgb, illuminant_RGB,
illuminant_XYZ, rgb_to_xyz_matrix,
chromatic_adaptation_transform)
lab = colour.XYZ_to_Lab(large_xyz, illuminant_XYZ)
return lab
def set_convertor(name, ill='D65'):
""" Binds the conversion functions LCH2RGB() and RGB2LCH() to the choosen colour package
"""
global LCH2RGB, RGB2LCH, convertor, illuminant
if name not in ['custom', 'colorspacious', 'colourscience']:
print("Unknown conversion module")
return
convertor = name
illuminant = ill
if name == 'custom':
LCH2RGB = lambda L, C, H: XYZ2RGB(Lab2XYZ(LCH2Lab((L, C, H))))
RGB2LCH = lambda R, G, B: Lab2LCH(XYZ2Lab(RGB2XYZ((R, G, B))))
if name == 'colorspacious':
from colorspacious import cspace_convert
func_LCH2RGB = lambda L, C, H: cspace_convert([L, C, H], {
"name": "CIELCh",
"XYZ100_w": ill
}, "sRGB1")
func_RGB2LCH = lambda R, G, B: cspace_convert([R, G, B], "sRGB1", {
"name": "CIELCh",
"XYZ100_w": ill
})
if name == 'colourscience':
import colour as cs
cs_ill = cs.ILLUMINANTS['CIE 1931 2 Degree Standard Observer'][ill]
func_LCH2RGB = lambda L, C, H: cs.XYZ_to_sRGB(
cs.Lab_to_XYZ(cs.LCHab_to_Lab([L, C, H]), illuminant=cs_ill))
func_RGB2LCH = lambda R, G, B: cs.Lab_to_LCHab(
cs.XYZ_to_Lab(cs.sRGB_to_XYZ([R, G, B]), illuminant=cs_ill))
if name == 'colorspacious' or name == 'colourscience':
def LCH2RGB(L, C, H):
if hasattr(L, '__iter__'):
RGB = np.array(list(map(func_LCH2RGB, L, C, H)))
R = RGB[:, 0]
G = RGB[:, 1]
B = RGB[:, 2]
else:
R, G, B = func_LCH2RGB(L, C, H)
return R, G, B
def RGB2LCH(R, G, B):
if hasattr(R, '__iter__'):
LCH = np.array(list(map(func_RGB2LCH, R, G, B)))
L = LCH[:, 0]
C = LCH[:, 1]
H = LCH[:, 2]
else:
L, C, H = func_RGB2LCH(R, G, B)
return L, C, H
print("convertor = '%s' (illuminant = '%s')" % (name, illuminant))
def ColorFidelityMetric(rgbX, rgbY):
M = 0
Q = 0
#W * H * 3に変形
rgbX_ = np.reshape(rgbX, (inputImg.shape[0], inputImg.shape[1], 3))
rgbY_ = np.reshape(rgbY, (inputImg.shape[0], inputImg.shape[1], 3))
#RGB-->Lab
XYZ_X = colour.sRGB_to_XYZ(rgbX_)
XYZ_Y = colour.sRGB_to_XYZ(rgbY_)
Lab_X = colour.XYZ_to_Lab(XYZ_X)
Lab_Y = colour.XYZ_to_Lab(XYZ_Y)
#ウィンドウサイズごとのFidelityMetricを計算
for (cropX, cropY) in SlidingWindow(Lab_X, Lab_Y, 1, 8):
# if the window does not meet our desired window size, ignore it
if cropX.shape[0] != 8 or cropX.shape[1] != 8:
continue
#オリジナル画像のクリップ
cropL_X = cropX[:, :, 0].flatten()
cropa_X = cropX[:, :, 1].flatten()
cropb_X = cropX[:, :, 2].flatten()
#加工画像のクリップ
cropL_Y = cropY[:, :, 0].flatten()
cropa_Y = cropY[:, :, 1].flatten()
cropb_Y = cropY[:, :, 2].flatten()
Ql = FidelityMetric(cropL_X, cropL_Y)
Qa = FidelityMetric(cropa_X, cropa_Y)
Qb = FidelityMetric(cropb_X, cropb_Y)
Q += np.sqrt(Ql**2 + Qa**2 + Qb**2)
M += 1
return Q / M
def BGR2LAB(image):
"""
Converts OpenCV BGR image to CIE Lab image
Lab Scale
L| 0:100 |
a| -100:100 |
b| -100:100 |
:param image: BGR image
:return: Lab image
"""
image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
image_srgb = image_rgb.astype(np.float32) / 255
image_lab = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(image_srgb))
return image_lab
def XYZ2LAB(xyz, illuminant='D65'):
return colour.XYZ_to_Lab(
xyz,
illuminant=colour.ILLUMINANTS['CIE 1931 2 Degree Standard Observer']
['D65'])
refwhite = [0.9504, 1.0000, 1.0888]
xyzr = np.array(xyz) / np.array(refwhite)
e = 0.008856
k = 903.3
ff = [0.0, 0.0, 0.0]
for i in range(0, 3):
if xyzr[i] > e:
ff[i] = xyzr[i]**(1.0 / 3.0)
else:
ff[i] = (k * xyzr[i] + 16) / 116
lab = [0.0, 0.0, 0.0]
lab[0] = ff[1] * 116 - 16
lab[1] = 500 * (ff[0] - ff[1])
lab[2] = 200 * (ff[1] - ff[2])
return lab
def get_main_color(R,G,B):
sRGB = [R/255,G/255,B/255]
RGB = np.array(sRGB)
XYZ = colour.sRGB_to_XYZ(sRGB)
Lab = colour.XYZ_to_Lab(XYZ)
LCHab = colour.Lab_to_LCHab(Lab)
L = int(LCHab[0]);C = int(LCHab[1]);ab = int(LCHab[2])
print([L,C,ab])
if C<40:# 1
if L<10:
return 'black'
elif L>90:
return 'white'
else:
return 'gray'
elif ab<45:# 2
if L<30:
return 'brown'
elif L>70:
return 'pink'
else:
return 'red'
elif ab<75:# 3
if L<30:
return 'brown'
else:
return 'orange'
elif ab<105:# 4
return 'yellow'
elif ab<210:# 5
return 'green'
elif ab<315:# 6
return 'blue'
elif ab>330:# 7
if L>50:
return 'pink'
else:
return 'purple'
else:
return 'purple'
def prepare_cc():
print(colour.COLOURCHECKERS.data.keys())
cc = []
for i in range(0, 24):
xyY = colour.COLOURCHECKERS.data['colorchecker 2005'][1].data[i].xyY
print(colour.COLOURCHECKERS.data['colorchecker 2005'][1].data[i].name)
print(xyY)
XYZ = colour.xyY_to_XYZ(xyY)
Lab = colour.XYZ_to_Lab(XYZ)
print(
colour.XYZ_to_sRGB(
XYZ, colour.ILLUMINANTS['CIE 1931 2 Degree Standard Observer']
['D50'], 'Bradford') * 255)
# print(colour.ILLUMINANTS['CIE 1931 2 Degree Standard Observer']['D50'])
XYZD65 = cam2xyz.XYZD50_XYZD65(XYZ)
cc.append(XYZD65)
print(XYZD65, ' ', colour.XYZ_to_sRGB(XYZD65) * 255)
# print(XYZ, ' ' ,XYZD65)
# print(colour.RGB_COLOURSPACES['sRGB'].whitepoint, ' ', ill2)
# print(colour.XYZ_to_sRGB(XYZD65) * 255)
# print(colour.COLOURCHECKERS.data['cc2005'][1].data[i])
return cc
def HEX_2_LAB(hex):
return colour.XYZ_to_Lab(
colour.sRGB_to_XYZ(colour.notation.HEX_to_RGB(hex)))
def Distance(xyz, lab):
lab1 = colour.XYZ_to_Lab(
xyz, colour.ILLUMINANTS['CIE 1931 2 Degree Standard Observer']['D65'])
return colour.delta_E(lab, lab1)
def time_4k(self):
colour.XYZ_to_Lab(xyz_4k)
def sRGB_to_Lab(r, g, b):
XYZ = colour.sRGB_COLOURSPACE.RGB_to_XYZ_matrix.dot(np.array([r, g, b]))
L, a, b = colour.XYZ_to_Lab(XYZ)
return L, a, b,
from __future__ import division, unicode_literals
import numpy as np
import colour
from colour.utilities.verbose import message_box
message_box('Overall "Colour" Examples')
message_box('N-Dimensional Arrays Support')
XYZ = (0.07049534, 0.10080000, 0.09558313)
illuminant = (0.34567, 0.35850)
message_box('Using 1d "array_like" parameter:\n' '\n{0}'.format(XYZ))
print(colour.XYZ_to_Lab(XYZ, illuminant=illuminant))
print('\n')
XYZ = np.tile(XYZ, (6, 1))
illuminant = np.tile(illuminant, (6, 1))
message_box('Using 2d "array_like" parameter:\n' '\n{0}'.format(XYZ))
print(colour.XYZ_to_Lab(XYZ, illuminant=illuminant))
print('\n')
XYZ = np.reshape(XYZ, (2, 3, 3))
illuminant = np.reshape(illuminant, (2, 3, 2))
message_box('Using 3d "array_like" parameter:\n' '\n{0}'.format(XYZ))
print(colour.XYZ_to_Lab(XYZ, illuminant=illuminant))
import colour
import numpy
import sys
targets = {}
with open('colors.data') as file:
for line in file:
code, name = line.strip().split(' ')
rgb = colour.notation.HEX_to_RGB(code)
targets[name] = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(rgb))
sources = {}
for arg in sys.argv[1:]:
rgb = colour.notation.HEX_to_RGB(arg)
sources[arg] = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(rgb))
mappings = {}
for arg, actual in sources.items():
diffs = []
for name, expected in targets.items():
distance = colour.delta_E(actual, expected)
diffs.append((distance, name))
mappings[arg] = sorted(diffs)
for arg, diffs in mappings.items():
for diff in diffs[:8]:
numpy.set_printoptions(precision=2)
print('{0} {2} {1:.2f}'.format(arg, *diff))
def update_CIELAB_3Dplot(fig, tristimulus_XYZ, selected_phi, data):
if fig is None or tristimulus_XYZ is None or selected_phi is None:
raise PreventUpdate
selected_phi = np.array(selected_phi)
phiVs = np.array(data['phiVs'])
thetaVs = np.array(data['thetaVs'])
tristimulus_XYZ = np.array(tristimulus_XYZ)
figure = go.Figure()
if selected_phi.shape[0] == 1:
selected_XYZ = tristimulus_XYZ[:, phiVs == selected_phi[0]][:, 0, :]
selected_LAB = np.array([
clr.XYZ_to_Lab(selected_XYZ[i] / 100)
for i in range(selected_XYZ.shape[0])
])
figure.add_trace(
go.Scatter(name='L*',
y=selected_LAB[:, 0],
x=thetaVs,
mode='lines+markers',
marker=dict(color='yellow'),
line=dict(color='yellow')))
figure.add_trace(
go.Scatter(name='a*',
y=selected_LAB[:, 1],
x=thetaVs,
mode='lines+markers',
marker=dict(color='red'),
line=dict(color='red')))
figure.add_trace(
go.Scatter(name='b*',
y=selected_LAB[:, 2],
x=thetaVs,
mode='lines+markers',
marker=dict(color='blue'),
line=dict(color='blue')))
elif selected_phi.shape[0] == 2:
selected_XYZ_pos = tristimulus_XYZ[:, phiVs == selected_phi[0]][:,
0, :]
selected_LAB_pos = np.array([
clr.XYZ_to_Lab(selected_XYZ_pos[i] / 100)
for i in range(selected_XYZ_pos.shape[0])
])
figure.add_trace(
go.Scatter(name='L* 0 to 90',
y=selected_LAB_pos[:, 0],
x=thetaVs,
mode='lines+markers',
marker=dict(color='yellow'),
line=dict(color='yellow')))
figure.add_trace(
go.Scatter(name='a* 0 to 90',
y=selected_LAB_pos[:, 1],
x=thetaVs,
mode='lines+markers',
marker=dict(color='red'),
line=dict(color='red')))
figure.add_trace(
go.Scatter(name='b* 0 to 90',
y=selected_LAB_pos[:, 2],
x=thetaVs,
mode='lines+markers',
marker=dict(color='blue'),
line=dict(color='blue')))
selected_XYZ_neg = tristimulus_XYZ[:, phiVs == selected_phi[1]][:,
0, :]
selected_XYZ_neg = np.array([
clr.XYZ_to_Lab(selected_XYZ_neg[i] / 100)
for i in range(selected_XYZ_neg.shape[0])
])
figure.add_trace(
go.Scatter(name='L* -90 to 0',
y=selected_XYZ_neg[:, 0],
x=-thetaVs,
mode='lines+markers',
marker=dict(color='yellow'),
line=dict(color='yellow')))
figure.add_trace(
go.Scatter(name='a* -90 to 0',
y=selected_XYZ_neg[:, 1],
x=-thetaVs,
mode='lines+markers',
marker=dict(color='red'),
line=dict(color='red')))
figure.add_trace(
go.Scatter(name='b* -90 to 0',
y=selected_XYZ_neg[:, 2],
x=-thetaVs,
mode='lines+markers',
marker=dict(color='blue'),
line=dict(color='blue')))
else:
raise PreventUpdate
figure.update_layout(
title="CIELab values dependence on viewing zenith angle",
xaxis_title='CIELab units',
yaxis_title='Viewing zenith angle Theta (deg)')
return figure
extend = np.array([
np.ones(24),
CameraXYZ[0], CameraXYZ[1], CameraXYZ[2],
CameraXYZ[0, :] * CameraXYZ[1, :],
CameraXYZ[1, :] * CameraXYZ[2, :],
CameraXYZ[2, :] * CameraXYZ[0, :],
CameraXYZ_2[0], CameraXYZ_2[1], CameraXYZ_2[2],
CameraXYZ[0, :] * CameraXYZ[1, :] * CameraXYZ[2, :],
CameraXYZ_3[0], CameraXYZ_3[1], CameraXYZ_3[2]
])
C = spectrumXYZ @ pinv(extend)
CorrectXYZ = C @ extend
Lab1 = colour.XYZ_to_Lab(CorrectXYZ.T, illuminant=colour.XYZ_to_xy(d65)) / 100
LabRGB11 = (colour.XYZ_to_sRGB(CameraXYZ.T / 100) * 255).astype(np.uint8)
Lab2 = colour.XYZ_to_Lab(spectrumXYZ.T / 100)
LabRGB21 = (colour.XYZ_to_sRGB(spectrumXYZ.T / 100) * 255).astype(np.uint8)
C_CIE76= np.sqrt(np.sum((Lab1 - Lab2) ** 2, 1))
C_AvgCIE76 = np.mean(C_CIE76)
C_CIE2000 = colour.delta_E_CIE2000(Lab1, Lab2).T
C_AvgCIE2000 = np.mean(C_CIE2000)
i = 24
RMSE_XYZ = np.sqrt(np.sum(CorrectXYZ[:, :i] - spectrumXYZ[:, :i]) ** 2 / 3)
RMSE_allXYZ = np.sum(RMSE_XYZ / 24)
coeff, scores, latent, explained = pca(color_Rspectrum.T, centered=False)
EV = coeff[:, :12]
im_xyz = raw.postprocess( #user_wb = user_wb,
use_camera_wb=True,
output_bps=16,
output_color=rp.ColorSpace.XYZ,
gamma=(1, 1)).astype(float) / (2.0**16 - 1)
# rgb = raw.postprocess(user_wb = user_wb)
# plt.imshow(rgb)
# plt.show()
fig, axs = plt.subplots()
dEs = np.zeros((4, 6), dtype=np.float)
for i in range(0, 24):
sample = im_xyz[box[i][1]:box[i][3], box[i][0]:box[i][2]]
mean_xyz = [np.mean(sample[:, :, x]) for x in range(0, 3)]
dE = colour.delta_E(
colour.XYZ_to_Lab(
mean_xyz,
illuminant=colour.
ILLUMINANTS['CIE 1931 2 Degree Standard Observer']['D65']),
colour.XYZ_to_Lab(
cc[i],
illuminant=colour.
ILLUMINANTS['CIE 1931 2 Degree Standard Observer']['D65']))
#print(dE)
dEs[int(i / 6)][i % 6] = dE
# vis = np.tile(mean_xyz, [120, 120, 1])
# vis2 = np.tile(cc[i], [120, 120, 1])
# fig, (ax1, ax2, ax3) = plt.subplots(1, 3)
# ax1.imshow(sample)
# ax2.imshow(vis)
# ax3.imshow(vis2)
# plt.show()
print(img.size) # 像素數量 print(img.dtype) # 數據類型 print(img) # 列印圖像的numpy數組，3緯數組 #%% # 網址1:https://colour.readthedocs.io/en/v0.3.13/generated/colour.sRGB_to_XYZ.html # 網址2:https://colour.readthedocs.io/en/v0.3.7/colour.html # 網址3:https://colour.readthedocs.io/en/v0.3.13/generated/colour.Lab_to_LCHab.html?highlight=lch#colour.Lab_to_LCHab # sRGB_to_XYZ #import colormath #from colormath.color_objects import LabColor, XYZColor #from colormath.color_conversions import convert_color import colour import numpy as np sRGB = [200 / 255, 150 / 255, 100 / 255] RGB = np.array(sRGB) XYZ = colour.sRGB_to_XYZ(sRGB) Lab = colour.XYZ_to_Lab(XYZ) LCHab = colour.Lab_to_LCHab(Lab) L = int(LCHab[0]) C = int(LCHab[1]) ab = int(LCHab[2]) #%% LCHab = ['41', '59', '27'] LCHab = np.array(LCHab) L = int(LCHab[0]) C = int(LCHab[1]) ab = int(LCHab[2]) #%% r = 255 g = 89 b = 0 sRGB = [r / 255, g / 255, b / 255]
def RGB_to_Lab(space, pt):
XYZ = colour.RGB_to_XYZ(pt,
space.whitepoint,
D50,
space.RGB_to_XYZ_matrix)
return colour.XYZ_to_Lab(XYZ)
print(colour.Luv_to_LCHuv(Luv))
print('\n')
LCHuv = (100.00000000, 70.90111393, 24.06324597)
message_box(('Converting to "CIE Luv" colourspace from given "CIE LCHuv" '
'colourspace values:\n'
'\n\t{0}'.format(LCHuv)))
print(colour.LCHuv_to_Luv(LCHuv))
print('\n')
message_box(('Converting to "CIE Lab" colourspace from given "CIE XYZ" '
'tristimulus values:\n'
'\n\t{0}'.format(XYZ)))
print(colour.XYZ_to_Lab(XYZ))
print('\n')
Lab = (100.00000000, 28.97832184, 30.96902832)
message_box(('Converting to "CIE XYZ" tristimulus values from given "CIE Lab" '
'colourspace values:\n'
'\n\t{0}'.format(Lab)))
print(colour.Lab_to_XYZ(Lab))
print('\n')
message_box(('Converting to "CIE LCHab" colourspace from given "CIE Lab" '
'colourspace values:\n'
'\n\t{0}'.format(Lab)))
print(colour.Lab_to_LCHab(Lab))
parser.add_argument('--alpha', type=float, default=1.0)
parser.add_argument('--gamma', type=float, default=1.0)
args = parser.parse_args()
print('load training data')
with gzip.open(args.training_data_path, 'rb') as f:
raw, jpeg = pickle.load(f)
raw = np.array(raw, dtype=np.float32)
jpeg = np.array(jpeg, dtype=np.float32)
raw /= 255
jpeg /= 255
if args.color_space == 'lab':
raw = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(raw))
jpeg = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(jpeg))
print('load xgboost models')
with open(args.model_file_path, 'rb') as f:
models = pickle.load(f)
print('lattice regression')
print('define A')
lattice_size = args.lut_size
lattice_points = np.linspace(0.0, 1.0, lattice_size)
A = np.array(
[p for p in product(lattice_points, lattice_points, lattice_points)],
"""
Showcases *Lightness* computations.
"""
import numpy as np
import colour
from colour.utilities.verbose import message_box
message_box('"Lightness" Computations')
xyY = (0.4316, 0.3777, 0.1008)
message_box(('Computing "Lightness" "CIE Lab" reference value for given '
'"CIE xyY" colourspace values:\n'
'\n\t{0}'.format(xyY)))
print(np.ravel(colour.XYZ_to_Lab(colour.xyY_to_XYZ(xyY)))[0])
print('\n')
Y = 10.08
message_box(('Computing "Lightness" using '
'"Glasser, Mckinney, Reilly and Schnelle (1958)" method for '
'given "luminance" value:\n'
'\n\t{0}'.format(Y)))
print(colour.lightness_Glasser1958(Y))
print(colour.lightness(Y, method='Glasser 1958'))
print('\n')
message_box(('Computing "Lightness" using "Wyszecki (1963)" method for '
'given "luminance" value:\n'
def compute(name, hue, lightness, saturation, output="scss"):
"""
Outputs different conversations of the given CIECAM02 input.
Uses CIECAM with the same "signature" as HLC
"""
xyz = colour.CIECAM02_to_XYZ(
lightness, saturation, hue,
WHITEPOINT_D65_2, adapting_luminance, background_luminance,
surround = surround,
discount_illuminant = discount_illuminant
)
xyz_trans = [ value / 100 for value in xyz ]
#lab_65_10 = colour.XYZ_to_Lab(xyz_trans, illuminant=ILLUMINANT_D65_10)
#lch_65_10 = colour.Lab_to_LCHab(lab_65_10)
#hlc_65_10 = [lch_65_10[2], lch_65_10[0], lch_65_10[1]]
lab_65_2 = colour.XYZ_to_Lab(xyz_trans, illuminant=ILLUMINANT_D65_2)
lch_65_2 = colour.Lab_to_LCHab(lab_65_2)
hlc_65_2 = [lch_65_2[2], lch_65_2[0], lch_65_2[1]]
#lab_50_10 = colour.XYZ_to_Lab(xyz_trans, illuminant=ILLUMINANT_D50_10)
#lch_50_10 = colour.Lab_to_LCHab(lab_50_10)
#hlc_50_10 = [lch_50_10[2], lch_50_10[0], lch_50_10[1]]
lab_50_2 = colour.XYZ_to_Lab(xyz_trans, illuminant=ILLUMINANT_D50_2)
lch_50_2 = colour.Lab_to_LCHab(lab_50_2)
hlc_50_2 = [lch_50_2[2], lch_50_2[0], lch_50_2[1]]
srgb = colour.XYZ_to_sRGB(xyz_trans)
validRGB = True
for entry in srgb:
if entry > 1:
validRGB = False
srgb_trans = [ value * 255 for value in srgb ] if validRGB else [0,0,0]
hex_trans = rgb_to_hex(srgb) if validRGB else "EXCEEDS"
if output == "text":
print("NAME :", name)
print("CAM-65 :", floatlist([hue, lightness, saturation]))
print("XYZ :", floatlist(xyz))
print("LAB-50/2° :", floatlist(lab_50_2))
print("HLC-50/2° :", floatlist(hlc_50_2))
print("LAB-65/2° :", floatlist(lab_65_2))
print("HLC-65/2° :", floatlist(hlc_65_2))
print("sRGB :", floatlist(srgb_trans))
print("HEX :", hex_trans)
print("")
elif output == "scss":
print("$%s: %s;" % (name, hex_trans))
elif output == "print":
print("%s = %s" % (name, intlist(lab_50_2)))
elif output == "affinity":
print("%s = %s" % (name, intlist(lab_65_2)))
print(colour.colorimetry.whiteness_Berger1959(XYZ, XYZ_0))
print("\n")
message_box(
f'Computing "whiteness" using "Taube (1960)" method for given sample and '
f'reference white "CIE XYZ" tristimulus values matrices:\n\n'
f"\t{XYZ}\n"
f"\t{XYZ_0}"
)
print(colour.whiteness(XYZ, XYZ_0, method="Taube 1960"))
print(colour.colorimetry.whiteness_Taube1960(XYZ, XYZ_0))
print("\n")
Lab = colour.XYZ_to_Lab(XYZ / 100, colour.XYZ_to_xy(XYZ_0 / 100))
message_box(
f'Computing "whiteness" using "Stensby (1968)" method for given sample '
f'"CIE L*a*b*" colourspace array:\n\n\t{Lab}'
)
print(colour.whiteness(XYZ, XYZ_0, method="Stensby 1968"))
print(colour.colorimetry.whiteness_Stensby1968(Lab))
print("\n")
message_box(
f'Computing "whiteness" using "ASTM E313" method for given sample '
f'"CIE XYZ" tristimulus values:\n\n\t{XYZ}'
)
print(colour.whiteness(XYZ, XYZ_0, method="ASTM E313"))
print(colour.colorimetry.whiteness_ASTME313(XYZ))
parser.add_argument('--color_space', type=str, default='lab')
parser.add_argument('--lut_size', type=int, default=33)
args = parser.parse_args()
print('load training data')
with gzip.open(args.training_data_path, 'rb') as f:
raw, jpeg = pickle.load(f)
raw = np.array(raw, dtype=np.float32)
jpeg = np.array(jpeg, dtype=np.float32)
raw /= 255
jpeg /= 255
if args.color_space == 'lab':
raw = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(raw))
jpeg = colour.XYZ_to_Lab(colour.sRGB_to_XYZ(jpeg))
learning_rate = 0.1
max_depth = 20
reg_lambda = 0.5
n_estimators = 100
print('train models')
models = []
for i in tqdm(range(3)):
if i == 0:
c = '(1, 0, 0)'
elif i == 1:
c = '(0, 1, 0)'
