import functools as ft
import numpy as np
import pyqtgraph as pg
from PyQt5.QtCore import QSize
from PyQt5.QtGui import QColor
from PyQt5.QtWidgets import QGridLayout, QPushButton, QSizePolicy, QWidget
from ClearMap.Visualization.Color import rand_cmap
[docs]
class LUTItem(pg.HistogramLUTItem):
"""Lookup table item for the DataViewer"""
def __init__(self, *args, **kwargs):
pg.HistogramLUTItem.__init__(self, *args, **kwargs)
self.vb.setMaximumWidth(15)
self.vb.setMinimumWidth(10)
[docs]
def imageChanged(self, autoLevel=False, autoRange=False):
if autoLevel:
mn, mx = self.quickMinMax(targetSize=500)
self.region.setRegion([mn, mx])
[docs]
def quickMinMax(self, targetSize=1e3):
"""
Estimate the min/max values of the image data by subsampling.
"""
data = self.imageItem().image
while data.size > targetSize:
ax = np.argmax(data.shape)
sl = [slice(None)] * data.ndim
sl[ax] = slice(None, None, 2)
data = data[tuple(sl)]
return np.nanmin(data), np.nanmax(data)
[docs]
class LUT(QWidget):
def __init__(self, image=None, color='red', percentiles=[[-100, 0, 50], [50, 75, 100]],
parent=None, *args):
QWidget.__init__(self, parent, *args)
self.layout = QGridLayout(self)
self.layout.setSpacing(0)
# self.layout.setMargin(0)
self.layout.setContentsMargins(0, 0, 0, 0)
self.lut = LUTWidget(parent=parent, image=image)
self.layout.addWidget(self.lut, 0, 0, 1, 1)
self.range_layout = QGridLayout()
self.range_buttons = []
pre = ['%d', '%d']
for r in range(2):
range_buttons_m = []
for i, p in enumerate(percentiles[r]):
button = QPushButton(pre[r] % (p))
button.setMaximumWidth(30)
font = button.font()
font.setPointSize(6)
button.setFont(font)
self.range_layout.addWidget(button, r, i)
range_buttons_m.append(button)
self.range_buttons.append(range_buttons_m)
self.layout.addLayout(self.range_layout, 1, 0, 1, 1)
self.precentiles = percentiles
self.percentile_id = [2, 2]
for m, ab in enumerate(self.range_buttons):
for p, abm in enumerate(ab):
abm.clicked.connect(ft.partial(self.updateRegionRange, m, p))
# default gradient
if color in pg.graphicsItems.GradientEditorItem.Gradients.keys():
self.lut.gradient.loadPreset(color)
elif color in pg.colormap.listMaps('matplotlib'):
colormap = pg.colormap.get(color, source='matplotlib')
self.lut.gradient.setColorMap(colormap)
elif color == 'random':
colormap_values = rand_cmap(int(image.image.max() - image.image.min()), map_type='bright', first_color_black=True, last_color_black=False)
colormap_values = [pg.mkColor(*[int(c*255) for c in col]) for col in colormap_values]
colormap = pg.ColorMap(None, colormap_values, mapping=pg.ColorMap.CLIP)
self.lut.gradient.setColorMap(colormap)
else:
self.lut.gradient.getTick(0).color = QColor(0, 0, 0, 0)
self.lut.gradient.getTick(1).color = QColor(color)
self.lut.gradient.updateGradient()
[docs]
def updateRegionRange(self, m, p):
self.percentile_id[m] = p
p_min = self.precentiles[0][self.percentile_id[0]]
p_max = self.precentiles[1][self.percentile_id[1]]
self.updateRegionPercentile(p_min, p_max)
[docs]
def updateRegionPercentile(self, pmin, pmax):
iitem = self.lut.imageItem()
if iitem is not None:
pmax1 = max(0, min(pmax, 100))
if pmin < 0:
pmin1 = min(-pmin, 100)
else:
pmin1 = min(pmin, 100)
if pmax1 == 0:
pmax1 = 1; pmax = 1
if pmin1 == 0:
pmin1 = 1; pmin = 1
r = [float(pmin)/pmin1, float(pmax)/pmax1] * self.quickPercentile(iitem.image, [pmin1, pmax1])
self.lut.region.setRegion(r)
[docs]
def quickPercentile(self, data, percentiles, target_size=1e3):
while data.size > target_size:
ax = np.argmax(data.shape)
sl = [slice(None)] * data.ndim
sl[ax] = slice(None, None, 2)
sl = tuple(sl)
data = data[sl]
if data.dtype == np.bool_:
return np.nanpercentile(data.astype(np.uint8), percentiles)
else:
return np.nanpercentile(data, percentiles)
