def genKaiser(neitz=True):
'''
'''
space = colorSpace()
kaiser = np.genfromtxt('data/kaiser1987.csv', delimiter=",")
# sort by subject:
subj1 = np.where(kaiser[:, 3] == 1)
subj2 = np.where(kaiser[:, 3] == 2)
s1_xy = kaiser[subj1, :2][0]
s2_xy = kaiser[subj2, :2][0]
# solve for z:
s1_z = 1.0 - (s1_xy[:, 0] + s1_xy[:, 1])
s2_z = 1.0 - (s2_xy[:, 0] + s2_xy[:, 1])
# combine:
s1_xyz = np.zeros((len(s1_xy), 3))
s1_xyz[:, :2] = s1_xy
s1_xyz[:, 2] = s1_z
s2_xyz = np.zeros((len(s2_xy), 3))
s2_xyz[:, :2] = s2_xy
s2_xyz[:, 2] = s2_z
jv = space._genJuddVos()
if neitz:
JV_Neitz_transMatrix = space._genJuddVos2Neitz(jv)
sub1_Neitz = np.dot(JV_Neitz_transMatrix, s1_xyz.T).T
sub2_Neitz = np.dot(JV_Neitz_transMatrix, s2_xyz.T).T
else:
sub1_Neitz = s1_xyz
sub2_Neitz = s2_xyz
return sub1_Neitz, sub2_Neitz, jv
def plotConeSpace():
'''
'''
space = colorSpace(fundamental='neitz',
LMSpeaks=[559.0, 530.0, 421.0])
space.plotColorSpace(space.Lnorm, space.Mnorm, space.spectrum)
plt.show()
def plotConfusionLines(deficit='tritan', clip=True):
'''add confusion lines
'''
space = colorSpace(fundamental='neitz', LMSpeaks=[559, 530, 417])
space.plotColorSpace()
space.find_copunctuals()
print deficit, ': ', space.copunctuals[deficit]
if deficit.lower() == 'deutan' or deficit.lower() == 'protan':
lambdas = [420, 460, 470, 480, 490, 500, 515,]
elif deficit.lower() == 'tritan':
lambdas = [420, 460, 480, 500, 520, 535, 545, 555,
570, 585, 600, 625, 700]
space.cs_ax.plot(space.copunctuals[deficit][0],
space.copunctuals[deficit][1], 'ko', markersize=8)
for lam in lambdas:
space.cs_ax.plot([space.find_testLightMatch(lam)[0],
space.copunctuals[deficit][0]],
[space.find_testLightMatch(lam)[1],
space.copunctuals[deficit][1]],
'k-', linewidth=1)
space.cs_ax.text(0.7, 1, deficit, fontsize=18)
if clip is True:
space.cs_ax.set_xlim([-0.4, 1.2])
space.cs_ax.set_ylim([-0.2, 1.2])
plt.show()
def plot_dichromatic_system(hybrid='ls', clip=True):
'''
'''
space = colorSpace(fundamental='neitz', LMSpeaks=[559, 530, 421])
space.plotColorSpace()
for x in np.arange(0, 1.1, 0.1):
if hybrid.lower() == 'ls' or hybrid.lower() == 'sl':
s = x
m = 0
l = -(1.0 - x)
elif hybrid.lower() == 'lm' or hybrid.lower() == 'ml':
s = 0
m = x
l = -(1.0 - x)
elif hybrid.lower() == 'ms' or hybrid.lower() == 'sm':
s = x
m = -(1.0 - x)
l = 0
else:
raise InputError('hybrid must be ls, lm or ms')
copunct = space.lms_to_rgb([l, m, s])
neutral_points = space.find_spect_neutral(copunct)
for neut in neutral_points:
space.cs_ax.plot([neut[0], copunct[0]], [neut[1], copunct[1]],
'-o', c=(np.abs(l), np.abs(m), np.abs(s)),
markersize=8, linewidth=2)
if clip is True:
space.cs_ax.set_xlim([-0.4, 1.2])
space.cs_ax.set_ylim([-0.2, 1.2])
plt.show()
def plotCIE():
'''
'''
space = colorSpace(fundamental='neitz',
LMSpeaks=[559.0, 530.0, 417.0])
sub1_Neitz, sub2_Neitz, jv = space.genKaiser()
## plot confusion lines
clip_area = Wedge((jv[0, 0], jv[0, 1]), r=10, theta1=0, theta2=360)
CIEcopunctuals = {'deutan': np.array([1.10, -0.1, 0.1]),
'protan': np.array([0.753, 0.247, 0]),
'tritan': np.array([0.17, 0, 0.83]),
}
for deficit in CIEcopunctuals:
space.plotColorSpace(rVal=jv[:, 0], gVal=jv[:, 1],
spec=space.spectrum)
print deficit, ': ', CIEcopunctuals[deficit]
if deficit.lower() == 'deutan' or deficit.lower() == 'protan':
lambdas = [420, 460, 470, 480, 490, 500, 515,]
elif deficit.lower() == 'tritan':
lambdas = [420, 460, 480, 500, 520, 535, 545, 555,
570, 585, 600, 625, 700]
space.cs_ax.plot(CIEcopunctuals[deficit][0],
CIEcopunctuals[deficit][1], 'ko', markersize=8)
for lam in lambdas:
R, G, B = jv[:, 0], jv[:, 1], jv[:, 2]
space.cs_ax.plot([space.find_testLightMatch(lam,
R, G, B)[0],
CIEcopunctuals[deficit][0]],
[space.find_testLightMatch(lam,
R, G, B)[1],
CIEcopunctuals[deficit][1]],
'k-', linewidth=1)
space.cs_ax.plot([jv[-1, 0], jv[0, 0]],
[jv[-1, 1], jv[0, 1]], 'k-', linewidth=3)
space.cs_ax.set_clip_path(clip_area)
space.cs_ax.set_ylim([-0.12, 0.9])
space.cs_ax.set_xlim([-0.05, 1.15])
space.cs_ax.set_xlabel('x', fontsize=10)
space.cs_ax.set_ylabel('y', fontsize=10)
space.cs_ax.text(0.8, 1, deficit, fontsize=18,
horizontalalignment='right',
verticalalignment='top',
transform=space.cs_ax.transAxes)
fig.set_tight_layout(True)
plt.show()
def plotCMFs():
'''
'''
space = colorSpace()
fig = plt.figure()
fig.set_tight_layout(True)
ax = fig.add_subplot(111)
pf.AxisFormat()
pf.TufteAxis(ax, ['left', 'bottom'], Nticks=[5, 5])
ax.plot(space.spectrum, space.CMFs[0, :], 'r', linewidth=2)
ax.plot(space.spectrum, space.CMFs[1, :], 'g', linewidth=2)
ax.plot(space.spectrum, space.CMFs[2, :], 'b', linewidth=2)
ax.set_xlim([space.spectrum[0], space.spectrum[-1]])
ax.set_xlabel('wavelength (nm)')
ax.set_ylabel('sensitivity')
plt.show()
def plotcoeff():
'''
'''
space = colorSpace()
fig = plt.figure()
fig.set_tight_layout(True)
ax = fig.add_subplot(111)
pf.AxisFormat()
pf.TufteAxis(ax, ['left', 'bottom'], Nticks=[5, 5])
ax.plot(space.spectrum, space.rVal, 'r', linewidth=2)
ax.plot(space.spectrum, space.gVal, 'g', linewidth=2)
ax.plot(space.spectrum, space.bVal, 'b', linewidth=2)
ax.set_xlim([space.spectrum[0], space.spectrum[-1]])
ax.set_xlabel('wavelength (nm)')
ax.set_ylabel('coefficients')
plt.show()
def plotLUV():
'''
'''
# make sure stim is cie 1931 and fundamentals neitz
space = colorSpace(fundamental='neitz',
LMSpeaks=[559.0, 530.0, 421.0],
stim='cie 1931')
space.genXYZ()
u = 4 * space.X / (-2 * space.X + 12 * space.Y + 3)
v = 9 * space.Y / (-2 * space.X + 12 * space.Y + 3)
ind1 = np.where(space.spectrum == 420)[0]
ind2 = np.where(space.spectrum == 700)[0]
spectrum = space.spectrum[ind1:ind2+1]
u = u[ind1:ind2]
v = v[ind1:ind2]
space.plotColorSpace(u, v, spectrum, ee=False, Luv=True,
skipLam=[530, 550])
space.cs_ax.axes.get_xaxis().set_visible(False)
space.cs_ax.axes.get_yaxis().set_visible(False)
plt.axis('off')
plt.show()
def plotRGsystem(PRINT=False, clip=True):
'''
'''
space = colorSpace(fundamental='neitz', LMSpeaks=[559, 530, 419])
space._plotColorSpace()
for l in range(0, 11):
m = (10.0 - l) / 10.0
l = l / 10.0
neut, RG = space.RG2lambda(0, m, l, True)
if PRINT is True:
#print RG
#print neut
print space.find_testlightFromRG(neut[0], neut[1])
space.cs_ax.plot([neut[0], RG[0]], [neut[1], RG[1]],
'-o', c=(l, m, 0), markersize=8, linewidth=2)
if clip is True:
space.cs_ax.set_xlim([-0.4, 1.2])
space.cs_ax.set_ylim([-0.2, 1.2])
plt.show()
def plotColorSpace(color):
'''
'''
space = colorSpace(fundamental='neitz', LMSpeaks=[559, 530, 421])
space.plotColorSpace(color=color)
plt.show()
def plotKaiser(neitz=True, showBY=True, clip=True,
showSub1=False, showSub2=True, stockman=False,
series=True):
'''
'''
space = colorSpace(fundamental='neitz',
LMSpeaks=[559.0, 530.0, 421.0])
sub1_Neitz, sub2_Neitz, jv = genKaiser()
if neitz:
space._plotColorSpace()
else:
space._plotColorSpace(rVal=jv[:, 0], gVal=jv[:, 1],
spec=space.spectrum)
space.cs_ax.plot([jv[-1, 0], jv[0, 0]],
[jv[-1, 1], jv[0, 1]], 'k-', linewidth=3)
space.cs_ax.set_ylim([0, 0.9])
space.cs_ax.set_xlim([-0.05, 0.8])
if showSub1:
space.cs_ax.plot(sub1_Neitz[:, 0], sub1_Neitz[:, 1], 'ko',
markersize=8, markeredgewidth=2,
markerfacecolor='w',
linewidth=2)
if showSub2:
space.cs_ax.plot(sub2_Neitz[:, 0], sub2_Neitz[:, 1], 'kx',
markersize=8, markeredgewidth=2, linewidth=2)
if showBY:
if stockman:
neut2, RG2 = space.BY2lambda(0, 0, 1., True)
c2 = (1, 0, 0)
c3 = (0.5, 0.5, 0)
neut3, RG3 = space.lambda2RG(522, False, True)
else:
neut2, RG2 = space.BY2lambda(1, 0, 0, True)
c2 = (0, 0, 1)
c3 = (0, 0.5, 0.5)
neut3, RG3 = space.lambda2BY(522, True)
neut1, RG1 = space.BY2lambda(0, 1., 0, True)
c1 = (0, 1, 0)
# plot green copunctual line
space.cs_ax.plot([neut1[0], RG1[0]], [neut1[1], RG1[1]],
'-o', c=c1, markersize=8, linewidth=2)
# plot red or blue copunctual depending on neitz or stockman
space.cs_ax.plot([neut2[0], RG2[0]], [neut2[1], RG2[1]],
'-o', c=c2, markersize=8, linewidth=2)
# plot
space.cs_ax.plot([neut3[0], RG3[0]], [neut3[1], RG3[1]],
'-o', c=c3, markersize=8, linewidth=2)
if stockman and series:
for lam in [500, 505, 510, 515]:
neut3, RG3 = space.lambda2RG(lam, False, True)
space.cs_ax.plot([neut3[0], RG3[0]], [neut3[1], RG3[1]],
'-o', c=c3, markersize=8, linewidth=2)
if clip is True:
space.cs_ax.set_xlim([-0.4, 1.2])
space.cs_ax.set_ylim([-0.2, 1.2])
space.cs_ax.set_xlabel('x', fontsize=10)
space.cs_ax.set_ylabel('y', fontsize=10)
plt.show()
import matplotlib.pylab as plt
import numpy as np
from base import plot as pf
from colorSpace import colorSpace
space = colorSpace()
def trichromaticAnalysis(Lmax=560, Smax=417):
'''
'''
M_lamMax = []
volume = []
for i in range(420, 561):
M_lamMax.append(i)
space.genLMS('Neitz', [Lmax, i, Smax])
space.setLights('stiles and burch')
space.genConvMatrix()
volume.append(np.linalg.det(space.convMatrix))
fig = plt.figure()
fig.set_tight_layout(True)
ax = fig.add_subplot(111)
pf.AxisFormat()
pf.TufteAxis(ax, ['left', 'bottom'], Nticks=[5, 5])
ax.plot(M_lamMax, volume, 'k', linewidth=3)
ind = np.where(np.array(M_lamMax) == 530)[0]
ax.plot(M_lamMax[ind], volume[ind], 'g+', markersize=15,
markeredgewidth=3)