def largeSeedWatershed(raw, pmap, seeds, membraneWidth = 7.0, visu=False):
blockShape = (100, )*3
cOpts = vbw.convOpts
pmap = numpy.require(pmap, dtype='float32')
with vigra.Timer("add noise"):
mx = pmap.max()
sshape = pmap.squeeze().shape
noise = numpy.random.rand(*sshape)*(0.05*mx)
noise = vigra.taggedView(noise.astype('float32'),'xyz')
opts = vbw.convOpts(blockShape=blockShape, sigma=4.0)
noise = vbw.gaussianSmooth(noise, options=opts)
pmap += noise
with vigra.Timer("smoothed tie breaker"):
# compute a smoothed map as tie breaker
opts = vbw.convOpts(blockShape=blockShape, sigma=membraneWidth/2.0)
gaussianSmoothedPmap = vbw.gaussianSmooth(pmap, options=opts)
addEps = 0.3
growingMap = gaussianSmoothedPmap
growingMap *= addEps
#
opts = vbw.convOpts(blockShape=blockShape, sigma=membraneWidth/7.50)
notSoMuch = vbw.gaussianSmooth(pmap, options=opts)
# get the actual growing mapz
growingMap += notSoMuch
growingMap /= 1.0 + addEps
with vigra.Timer("watershedsNew"):
# do the actual watershed
growingMap = vigra.taggedView(growingMap, 'xyz')
seeds = numpy.require(seeds, dtype='uint32')
seeds = vigra.taggedView(seeds, 'xyz')
seg,nSeg = vigra.analysis.watershedsNew(image=growingMap, seeds=seeds)
if visu:
grayData = [
(raw, "raw"),
(pmap,"pmap"),
(growingMap,"growingMap")
]
segData = [
(seeds, "seeds"),
(seg, "seg")
]
skneuro.addHocViewer(grayData, segData)
return seg,nSeg
def largeSeedWatershed(raw, pmap, seeds, membraneWidth=7.0, visu=False):
blockShape = (100, ) * 3
cOpts = vbw.convOpts
pmap = numpy.require(pmap, dtype='float32')
with vigra.Timer("add noise"):
mx = pmap.max()
sshape = pmap.squeeze().shape
noise = numpy.random.rand(*sshape) * (0.05 * mx)
noise = vigra.taggedView(noise.astype('float32'), 'xyz')
opts = vbw.convOpts(blockShape=blockShape, sigma=4.0)
noise = vbw.gaussianSmooth(noise, options=opts)
pmap += noise
with vigra.Timer("smoothed tie breaker"):
# compute a smoothed map as tie breaker
opts = vbw.convOpts(blockShape=blockShape, sigma=membraneWidth / 2.0)
gaussianSmoothedPmap = vbw.gaussianSmooth(pmap, options=opts)
addEps = 0.3
growingMap = gaussianSmoothedPmap
growingMap *= addEps
#
opts = vbw.convOpts(blockShape=blockShape, sigma=membraneWidth / 7.50)
notSoMuch = vbw.gaussianSmooth(pmap, options=opts)
# get the actual growing mapz
growingMap += notSoMuch
growingMap /= 1.0 + addEps
with vigra.Timer("watershedsNew"):
# do the actual watershed
growingMap = vigra.taggedView(growingMap, 'xyz')
seeds = numpy.require(seeds, dtype='uint32')
seeds = vigra.taggedView(seeds, 'xyz')
seg, nSeg = vigra.analysis.watershedsNew(image=growingMap, seeds=seeds)
if visu:
grayData = [(raw, "raw"), (pmap, "pmap"), (growingMap, "growingMap")]
segData = [(seeds, "seeds"), (seg, "seg")]
skneuro.addHocViewer(grayData, segData)
return seg, nSeg
from vigra import blockwise as bw
numpy.random.seed(42)
# input
shape = (500, 500, 500)
data = numpy.random.rand(*shape).astype('float32')
print "make options object"
options = bw.BlockwiseConvolutionOptions3D()
print type(options)
sigma = 1.0
options.stdDev = (sigma, )*3
options.blockShape = (128, )*3
print "stddev",options.stdDev
print "call blockwise filter"
with vigra.Timer("AllThread"):
res = bw.gaussianSmooth(data, options)
with vigra.Timer("1thread"):
resRef = vigra.gaussianSmoothing(data, sigma)
print numpy.sum(numpy.abs(res-resRef))
def onClickedComputeFeaturesImpl(self, param):
lCtrl = self.layerCtrl
extractor = self.featureExtractor
rawData = self.dataDict['raw']
print "compute features"
print param
options = bw.BlockwiseConvolutionOptions3D()
options.blockShape = (128, )*3
fSize = [0]
def fRange(feat):
old = long(fSize[0])
fSize[0] += feat.shape[1]
return (old, fSize[0])
geoFeat,geoFeatNames = extractor.geometricFeatures()
lCtrl.addFeature("GeometricFeature",fRange(geoFeat),subNames=geoFeatNames)
topoFeat, topoFeatNames = extractor.topologicalFeatures()
lCtrl.addFeature("TopologicalFeatures",fRange(topoFeat),subNames=topoFeatNames)
features = [geoFeat, topoFeat]
for order in range(3):
orderName = '%d-Order Filter'%order
doFilter = param[('RawData',orderName,'computeFilter')]
if(doFilter):
sigmas = param[('RawData',orderName,'sigma')]
sigmas = list(eval(sigmas))
doUcm = param[('RawData',orderName,'UCM','ucmFilters')]
ucmMeanSign = float(param[('RawData',orderName,'UCM','meanSign')])
print "ucmMeanSign",ucmMeanSign
print "doUcm",doUcm
wardness = param[('RawData',orderName,'UCM','wardness')]
wardness = list(eval(wardness))
wardness = numpy.array(wardness, dtype='float32')
wnames =[]
for w in wardness :
wnames.append(" w"+str(w))
wnames.append(" r"+str(w))
print "sigma",sigmas, "w",wardness
for sigma in sigmas:
print sigma
orderSigmaName = orderName + str(sigma)
options.stdDev = (float(sigma), )*3
if order == 0:
if sigma < 0.01:
res = rawData
else:
res = bw.gaussianSmooth(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature(orderSigmaName,fRange(accFeat),subNames=accFeatNames)
features.append(accFeat)
if doUcm :
mean = accFeat[:,0]
mean *= ucmMeanSign
ucmName = orderSigmaName+"MeanUcm"
ucm = extractor.ucmTransformFeatures(mean[:,None], wardness)
lCtrl.addFeature(ucmName,fRange(ucm),subNames=wnames)
features.append(ucm)
if order == 1 :
if sigma < 0.5:
continue
else:
res = bw.gaussianGradientMagnitude(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature(orderSigmaName,fRange(accFeat),subNames=accFeatNames)
features.append(accFeat)
if doUcm :
mean = accFeat[:,0]
print "ucmMeanSign",ucmMeanSign
mean *= ucmMeanSign
ucmName = orderSigmaName+"MeanUcm"
ucm = extractor.ucmTransformFeatures(mean[:,None], wardness)
lCtrl.addFeature(ucmName,fRange(ucm),subNames=wnames)
features.append(ucm)
if order == 2 :
if sigma < 0.5:
continue
else:
res = bw.hessianOfGaussianFirstEigenvalue(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature(orderSigmaName,fRange(accFeat),subNames=accFeatNames)
features.append(accFeat)
if doUcm :
mean = accFeat[:,0]
mean *= ucmMeanSign
ucmName = orderSigmaName+"MeanUcm"
ucm = extractor.ucmTransformFeatures(mean[:,None], wardness)
lCtrl.addFeature(ucmName,fRange(ucm),subNames=wnames)
features.append(ucm)
if False:
for s in [2.0]:
options.stdDev = (s, )*3
res = bw.gaussianSmooth(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature("RawGaussianSmooth",fRange(accFeat),subNames=accFeatNames)
features.append(accFeat)
# 6
wardness = numpy.array([0.0, 0.1, 0.15, 0.25, 0.5, 1.0], dtype='float32')
wnames =[]
for w in wardness :
wnames.append(" w"+str(w))
wnames.append(" r"+str(w))
for s in [2.0]:
options.stdDev = (s, )*3
res = bw.hessianOfGaussianFirstEigenvalue(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature("hessianOfGaussianFirstEigenvalue",fRange(accFeat),subNames=accFeatNames)
features.append(accFeat)
print "ucm"
mean = accFeat[:,0]
ucm = extractor.ucmTransformFeatures(mean[:,None], wardness)
print "ucm done"
lCtrl.addFeature("hessianMeanUcm",fRange(ucm),subNames=wnames)
features.append(ucm)
#for s in [1.0, 3.0, 4.0]:
# img = hessianEv2(rawData, s, 2.0)
# #vigra.imshow(img)
# #vigra.show()
# accFeat = extractor.accumulatedFeatures(img)
# features.append(accFeat)
# mean = accFeat[:,0]
# ucm = extractor.ucmTransformFeatures(mean[:,None],wardness)
# features.extend([accFeat,ucm])
features = numpy.concatenate(features,axis=1)
self.currentFeatures = features
print "compute done"
fMin = numpy.min(features, axis=0)
fMax = numpy.max(features, axis=0)
self.featureMinMax = (fMin, fMax)
self.ctrlWidget.setFeatures(self.currentFeatures.shape[1])
self.ctrlWidget.modeSelectorComboBox.setCurrentIndex(2)
if True:
if True:
options = bw.BlockwiseConvolutionOptions3D()
sigma = 2.7
options.stdDev = (sigma, )*3
options.blockShape = (64, )*3
print "hessianEv"
ev = bw.hessianOfGaussianFirstEigenvalue(volume, options)
print ev.shape, ev.dtype
options.stdDev = (4.5, )*3
print "smooth"
ev = bw.gaussianSmooth(ev, options)
vigra.impex.writeHDF5(ev,"growmap.h5",'data')
else:
ev = vigra.impex.readHDF5("growmap.h5",'data')
with vigra.Timer("watershedsNew"):
labels, nseg = vigra.analysis.unionFindWatershed3D(ev,(100,100,100))
print "gridGraph"
gridGraph = graphs.gridGraph(labels.shape)
rag = graphs.regionAdjacencyGraph(gridGraph, labels)
rag.writeHDF5("rag.h5",'data')
else:
import vigra
from vigra import graphs
from vigra import numpy
from vigra import Timer
from vigra import blockwise as bw
numpy.random.seed(42)
# input
shape = (500, 500, 500)
data = numpy.random.rand(*shape).astype('float32')
print "make options object"
options = bw.BlockwiseConvolutionOptions3D()
print type(options)
sigma = 1.0
options.stdDev = (sigma, ) * 3
options.blockShape = (128, ) * 3
print "stddev", options.stdDev
print "call blockwise filter"
with vigra.Timer("AllThread"):
res = bw.gaussianSmooth(data, options)
with vigra.Timer("1thread"):
resRef = vigra.gaussianSmoothing(data, sigma)
print numpy.sum(numpy.abs(res - resRef))
def onClickedComputeFeaturesImpl(self, param):
lCtrl = self.layerCtrl
extractor = self.featureExtractor
rawData = self.dataDict['raw']
print "compute features"
print param
options = bw.BlockwiseConvolutionOptions3D()
options.blockShape = (128, ) * 3
fSize = [0]
def fRange(feat):
old = long(fSize[0])
fSize[0] += feat.shape[1]
return (old, fSize[0])
geoFeat, geoFeatNames = extractor.geometricFeatures()
lCtrl.addFeature("GeometricFeature",
fRange(geoFeat),
subNames=geoFeatNames)
topoFeat, topoFeatNames = extractor.topologicalFeatures()
lCtrl.addFeature("TopologicalFeatures",
fRange(topoFeat),
subNames=topoFeatNames)
features = [geoFeat, topoFeat]
for order in range(3):
orderName = '%d-Order Filter' % order
doFilter = param[('RawData', orderName, 'computeFilter')]
if (doFilter):
sigmas = param[('RawData', orderName, 'sigma')]
sigmas = list(eval(sigmas))
doUcm = param[('RawData', orderName, 'UCM', 'ucmFilters')]
ucmMeanSign = float(param[('RawData', orderName, 'UCM',
'meanSign')])
print "ucmMeanSign", ucmMeanSign
print "doUcm", doUcm
wardness = param[('RawData', orderName, 'UCM', 'wardness')]
wardness = list(eval(wardness))
wardness = numpy.array(wardness, dtype='float32')
wnames = []
for w in wardness:
wnames.append(" w" + str(w))
wnames.append(" r" + str(w))
print "sigma", sigmas, "w", wardness
for sigma in sigmas:
print sigma
orderSigmaName = orderName + str(sigma)
options.stdDev = (float(sigma), ) * 3
if order == 0:
if sigma < 0.01:
res = rawData
else:
res = bw.gaussianSmooth(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(
res)
lCtrl.addFeature(orderSigmaName,
fRange(accFeat),
subNames=accFeatNames)
features.append(accFeat)
if doUcm:
mean = accFeat[:, 0]
mean *= ucmMeanSign
ucmName = orderSigmaName + "MeanUcm"
ucm = extractor.ucmTransformFeatures(
mean[:, None], wardness)
lCtrl.addFeature(ucmName,
fRange(ucm),
subNames=wnames)
features.append(ucm)
if order == 1:
if sigma < 0.5:
continue
else:
res = bw.gaussianGradientMagnitude(
rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(
res)
lCtrl.addFeature(orderSigmaName,
fRange(accFeat),
subNames=accFeatNames)
features.append(accFeat)
if doUcm:
mean = accFeat[:, 0]
print "ucmMeanSign", ucmMeanSign
mean *= ucmMeanSign
ucmName = orderSigmaName + "MeanUcm"
ucm = extractor.ucmTransformFeatures(
mean[:, None], wardness)
lCtrl.addFeature(ucmName,
fRange(ucm),
subNames=wnames)
features.append(ucm)
if order == 2:
if sigma < 0.5:
continue
else:
res = bw.hessianOfGaussianFirstEigenvalue(
rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(
res)
lCtrl.addFeature(orderSigmaName,
fRange(accFeat),
subNames=accFeatNames)
features.append(accFeat)
if doUcm:
mean = accFeat[:, 0]
mean *= ucmMeanSign
ucmName = orderSigmaName + "MeanUcm"
ucm = extractor.ucmTransformFeatures(
mean[:, None], wardness)
lCtrl.addFeature(ucmName,
fRange(ucm),
subNames=wnames)
features.append(ucm)
if False:
for s in [2.0]:
options.stdDev = (s, ) * 3
res = bw.gaussianSmooth(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature("RawGaussianSmooth",
fRange(accFeat),
subNames=accFeatNames)
features.append(accFeat)
# 6
wardness = numpy.array([0.0, 0.1, 0.15, 0.25, 0.5, 1.0],
dtype='float32')
wnames = []
for w in wardness:
wnames.append(" w" + str(w))
wnames.append(" r" + str(w))
for s in [2.0]:
options.stdDev = (s, ) * 3
res = bw.hessianOfGaussianFirstEigenvalue(rawData, options)
accFeat, accFeatNames = extractor.accumulatedFeatures(res)
lCtrl.addFeature("hessianOfGaussianFirstEigenvalue",
fRange(accFeat),
subNames=accFeatNames)
features.append(accFeat)
print "ucm"
mean = accFeat[:, 0]
ucm = extractor.ucmTransformFeatures(mean[:, None], wardness)
print "ucm done"
lCtrl.addFeature("hessianMeanUcm",
fRange(ucm),
subNames=wnames)
features.append(ucm)
#for s in [1.0, 3.0, 4.0]:
# img = hessianEv2(rawData, s, 2.0)
# #vigra.imshow(img)
# #vigra.show()
# accFeat = extractor.accumulatedFeatures(img)
# features.append(accFeat)
# mean = accFeat[:,0]
# ucm = extractor.ucmTransformFeatures(mean[:,None],wardness)
# features.extend([accFeat,ucm])
features = numpy.concatenate(features, axis=1)
self.currentFeatures = features
print "compute done"
fMin = numpy.min(features, axis=0)
fMax = numpy.max(features, axis=0)
self.featureMinMax = (fMin, fMax)
self.ctrlWidget.setFeatures(self.currentFeatures.shape[1])
self.ctrlWidget.modeSelectorComboBox.setCurrentIndex(2)
if True:
if True:
options = bw.BlockwiseConvolutionOptions3D()
sigma = 2.7
options.stdDev = (sigma, ) * 3
options.blockShape = (64, ) * 3
print "hessianEv"
ev = bw.hessianOfGaussianFirstEigenvalue(volume, options)
print ev.shape, ev.dtype
options.stdDev = (4.5, ) * 3
print "smooth"
ev = bw.gaussianSmooth(ev, options)
vigra.impex.writeHDF5(ev, "growmap.h5", 'data')
else:
ev = vigra.impex.readHDF5("growmap.h5", 'data')
with vigra.Timer("watershedsNew"):
labels, nseg = vigra.analysis.unionFindWatershed3D(ev, (100, 100, 100))
print "gridGraph"
gridGraph = graphs.gridGraph(labels.shape)
rag = graphs.regionAdjacencyGraph(gridGraph, labels)
rag.writeHDF5("rag.h5", 'data')
else:
rag = vigra.graphs.loadGridRagHDF5("rag.h5", 'data')