def getGroundTruthWatershed(groundTruth, threshold=120):
groundTruthThreshold = copy.deepcopy(groundTruth)
groundTruthThreshold[groundTruthThreshold < threshold] = 0
groundTruthThreshold[groundTruthThreshold >= threshold] = 255
groundTruthLabels = vigra.analysis.labelImageWithBackground(groundTruthThreshold)
### Remove Mini-Label-Areas
for i in range(1, groundTruthLabels.max()+1):
if (len(groundTruthLabels[groundTruthLabels==i]) < 4):
groundTruthLabels[groundTruthLabels==i] = 0
groundTruthLabels = vigra.analysis.labelImageWithBackground(groundTruthLabels)
numLabels = groundTruthLabels.max()
groundTruthCont = copy.deepcopy(groundTruthThreshold)
groundTruthCont[groundTruthCont==0] = 1
groundTruthCont[groundTruthCont==255] = 0
groundTruthCont = vigra.filters.gaussianSmoothing(groundTruthCont, 0.1)
seeds = np.zeros((groundTruth.shape[0],groundTruth.shape[1]), dtype=np.uint32)
for i in range(1, numLabels+1):
labelArea = np.where(groundTruthLabels==i)
x = labelArea[0][0]
y = labelArea[1][0]
seeds[x][y] = i
watershedLabels = np.transpose(graphs.nodeWeightedWatersheds(rag.baseGraph, groundTruthCont, seeds))
return watershedLabels
def testGridGraphWatersheds():
data = numpy.random.random([10, 10, 10]).astype(numpy.float32)
edata = numpy.random.random([10 * 2 - 1, 10 * 2 - 1,
10 * 2 - 1]).astype(numpy.float32)
g0 = graphs.gridGraph(data.shape)
ew = graphs.edgeFeaturesFromInterpolatedImage(graph=g0, image=edata)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0, nodeWeights=data)
# node weighted watershed seeds
labelsNodeWeightedA = graphs.nodeWeightedWatersheds(graph=g0,
nodeWeights=data,
seeds=seeds)
# node weighted watershed seeds
labelsNodeWeightedB = graphs.nodeWeightedWatersheds(graph=g0,
nodeWeights=data)
# edge weighted watershed seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0, nodeWeights=data)
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(graph=g0,
edgeWeights=ew,
seeds=seeds)
assert numpy.array_equal(labelsNodeWeightedA, labelsNodeWeightedB)
data = numpy.random.random([10, 10]).astype(numpy.float32)
edata = numpy.random.random([10 * 2 - 1, 10 * 2 - 1]).astype(numpy.float32)
g0 = graphs.gridGraph(data.shape)
ew = graphs.edgeFeaturesFromInterpolatedImage(graph=g0, image=edata)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0, nodeWeights=data)
# node weighted watershed seeds
labelsNodeWeightedA = graphs.nodeWeightedWatersheds(graph=g0,
nodeWeights=data,
seeds=seeds)
# node weighted watershed seeds
labelsNodeWeightedB = graphs.nodeWeightedWatersheds(graph=g0,
nodeWeights=data)
# edge weighted watershed seeds
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(graph=g0,
edgeWeights=ew,
seeds=seeds)
assert numpy.array_equal(labelsNodeWeightedA, labelsNodeWeightedB)
def testGridGraphWatersheds():
data = numpy.random.random([10,10,10]).astype(numpy.float32)
edata = numpy.random.random([10*2-1,10*2-1,10*2-1]).astype(numpy.float32)
g0 = graphs.gridGraph(data.shape)
ew = graphs.edgeFeaturesFromInterpolatedImage(graph=g0,image=edata)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0,nodeWeights=data)
# node weighted watershed seeds
labelsNodeWeightedA = graphs.nodeWeightedWatersheds(graph=g0,nodeWeights=data,seeds=seeds)
# node weighted watershed seeds
labelsNodeWeightedB = graphs.nodeWeightedWatersheds(graph=g0,nodeWeights=data)
# edge weighted watershed seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0,nodeWeights=data)
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(graph=g0,edgeWeights=ew,seeds=seeds)
assert numpy.array_equal(labelsNodeWeightedA,labelsNodeWeightedB)
data = numpy.random.random([10,10]).astype(numpy.float32)
edata = numpy.random.random([10*2-1,10*2-1]).astype(numpy.float32)
g0 = graphs.gridGraph(data.shape)
ew = graphs.edgeFeaturesFromInterpolatedImage(graph=g0,image=edata)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(graph=g0,nodeWeights=data)
# node weighted watershed seeds
labelsNodeWeightedA = graphs.nodeWeightedWatersheds(graph=g0,nodeWeights=data,seeds=seeds)
# node weighted watershed seeds
labelsNodeWeightedB = graphs.nodeWeightedWatersheds(graph=g0,nodeWeights=data)
# edge weighted watershed seeds
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(graph=g0,edgeWeights=ew,seeds=seeds)
assert numpy.array_equal(labelsNodeWeightedA,labelsNodeWeightedB)
gridGraphEdgeIndicator = graphs.edgeFeaturesFromInterpolatedImage(gridGraph,
gradMagBig)
# get region adjacency graph from super-pixel labels
rag = graphs.regionAdjacencyGraph(gridGraph, labels)
# accumulate edge and ndie weights from gradient magnitude
ragEdgeWeights = rag.accumulateEdgeFeatures(gridGraphEdgeIndicator)
ragNodeWeights = rag.accumulateNodeFeatures(gradMag)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(rag, ragNodeWeights)
# node weighted watersheds
labelsNodeWeighted = graphs.nodeWeightedWatersheds(rag, ragNodeWeights, seeds)
# edge weighted watersheds
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(rag, ragEdgeWeights, seeds)
f = pylab.figure()
ax0 = f.add_subplot(1, 2, 0)
rag.showNested(img, labelsNodeWeighted)
ax0.set_title("node weighted")
ax1 = f.add_subplot(1, 2, 1)
rag.showNested(img, labelsEdgeWeighted)
ax1.set_title("edge weighted")
pylab.show()
gridGraphEdgeIndicator = graphs.edgeFeaturesFromInterpolatedImage(gridGraph,
gradMagBig)
# get region adjacency graph from super-pixel labels
rag = graphs.regionAdjacencyGraph(gridGraph, labels)
# accumulate edge and ndie weights from gradient magnitude
ragEdgeWeights = rag.accumulateEdgeFeatures(gridGraphEdgeIndicator)
ragNodeWeights = rag.accumulateNodeFeatures(gradMag)
# generate seeds
seeds = graphs.nodeWeightedWatershedsSeeds(rag, ragNodeWeights)
# node weighted watersheds
labelsNodeWeighted = graphs.nodeWeightedWatersheds(rag, ragNodeWeights, seeds)
# edge weighted watersheds
labelsEdgeWeighted = graphs.edgeWeightedWatersheds(rag, ragEdgeWeights, seeds)
f = pylab.figure()
ax0 = f.add_subplot(1, 2, 0)
rag.showNested(img, labelsNodeWeighted)
ax0.set_title("node weighted")
ax1 = f.add_subplot(1, 2, 1)
rag.showNested(img, labelsEdgeWeighted)
ax1.set_title("edge weighted")
pylab.show()
