def absdiff(imgA, imgB):
return let(
"diff",
sub(imgA, imgB),
IF(
LT("diff", 0),
THEN(sub(0, "diff")), # invert the sign to make it positive
ELSE("diff")))
def findPeaks(img4D, params):
"""
img4D: a 4D RandomAccessibleInterval
params["frames"]: the number of consecutive time points to average
towards detecting peaks with difference of Gaussian.
Returns a list of lists of peaks found, one list per time point.
"""
frames = params["frames"]
# Work image: the current sum
sum3D = ArrayImgs.unsignedLongs([img4D.dimension(d) for d in [0, 1, 2]])
peaks = []
# Sum of the first set of frames
compute(add([Views.hyperSlice(img4D, 3, i) for i in xrange(frames)])).into(sum3D)
# Extract nuclei from first sum3D
peaks.append(doGPeaks(sum3D, params))
# Running sums: subtract the first and add the last
for i in xrange(frames, img4D.dimension(3), 1):
compute(add(sub(sum3D,
Views.hyperSlice(img4D, 3, i - frames)),
Views.hyperSlice(img4D, 3, i))) \
.into(sum3D)
# Extract nuclei from sum4D
peaks.append(doGPeaks(sum3D, params))
return peaks
def filterBankBlockStatistics(img, block_width=5, block_height=5,
integralImgE=None,
sumType=UnsignedLongType(),
converter=Util.genericRealTypeConverter()):
# corners of a block centered at the pixel
block_width = int(block_width)
block_height = int(block_height)
w0 = -block_width/2 # e.g. -2 when block_width == 4 and also when block_width == 5
h0 = -block_height/2
decX = 1 if 0 == block_width % 2 else 0
decY = 1 if 0 == block_height % 2 else 0
w1 = block_width/2 - decX # e.g. 2 when block_width == 5 but 1 when block_width == 4
h1 = block_height/2 - decY
corners = [[w0, h0], [w1, h0],
[w0, h1], [w1, h1]]
# Create the integral image, stored as 64-bit
if not integralImgE:
alg = IntegralImg(img, sumType, converter)
alg.process()
integralImgE = Views.extendBorder(alg.getResult())
# Create the integral image of squares, stored as 64-bit
sqimg = compute(power(img, 2)).view(sumType)
algSq = IntegralImg(sqimg, sumType, converter)
algSq.process()
integralImgSqE = Views.extendBorder(algSq.getResult())
# block mean: creates holes in blurred membranes
opMean = div(block(integralImgSqE, corners), block_width * block_height)
# block variance: sum of squares minus square of sum
opVariance = sub(block(integralImgSqE, corners), power(block(integralImgE, corners), 2))
opVarianceNonZero = let("var", opVariance,
IF(LT("var", 0),
THEN(0),
ELSE("var")))
return [opMean, opVarianceNonZero]
# Compute the brightness: pick the maximum intensity pixel of every channel
# and then normalize it by dividing by the number of channels
compute(div(maximum([red, green, blue]), 255.0)).into(brightness)
# Show the brightness image
impB = IL.wrap(brightness, imp.getTitle() + " brightness")
impB.show()
# Compute now the image color saturation
saturation = ArrayImgFactory(FloatType()).create(img)
compute(
let(
"red",
red, # store as directly readable variables (no dictionary lookups)
"green",
green, # so that only 3 cursors are needed instead of 6
"blue",
blue,
"max",
maximum([var("red"), var("green"),
var("blue")]),
"min",
minimum([var("red"), var("green"),
var("blue")]),
IF(EQ(0, var("max")), THEN(0),
ELSE(div(sub(var("max"), var("min")),
var("max")))))).into(saturation)
impC = IL.wrap(saturation, imp.getTitle() + " saturation")
impC.show()
# Create the integral image, stored as 64-bit alg = IntegralImg(img, UnsignedLongType(), Util.genericIntegerTypeConverter()) alg.process() integralImg = alg.getResult() imgE = Views.extendBorder(integralImg) # Haar-like features from Viola and Jones # tuned to identify neuron membranes # Two adjacent vertical rectangles 4x8 - 4x8 cornersVL = [[-3, -4], [0, -4], [-3, 3], [0, 3]] cornersVR = [[1, -4], [4, -4], [1, 3], [4, 3]] blockVL = block(imgE, cornersVL) blockVR = block(imgE, cornersVR) op1 = sub(blockVL, blockVR) op2 = sub(blockVR, blockVL) # Two adjacent horizontal rectangles 4x8 - 4x8 cornersHT = [[-4, -3], [3, -3], [-4, 0], [3, 0]] cornersHB = [[-4, 1], [3, 1], [-4, 4], [3, 4]] blockHT = block(imgE, cornersHT) blockHB = block(imgE, cornersHB) op3 = sub(blockHT, blockHB) op4 = sub(blockHB, blockHT) # Two bright-black-bright vertical features 4x8 - 4x8 - 4x8 corners3VL = [[x - 2, y] for x, y in cornersVL] corners3VC = [[x + 2, y] for x, y in cornersVL] corners3VR = [[x + 2, y] for x, y in cornersVR] print corners3VL
def filterBank(img, bs=4, bl=8, sumType=UnsignedLongType(), converter=Util.genericRealTypeConverter()):
""" Haar-like features from Viola and Jones using integral images
tuned to identify neuron membranes in electron microscopy.
bs: length of the short side of a block
bl: length of the long side of a block
sumType: the type with which to add up pixel values in the integral image
converter: for the IntegralImg, to convert from input to sumType
"""
# Create the integral image, stored as 64-bit
alg = IntegralImg(img, sumType, converter)
alg.process()
integralImg = alg.getResult()
imgE = Views.extendBorder(integralImg)
# corners of a 4x8 or 8x4 rectangular block where 0,0 is the top left
cornersV = [[0, 0], [bs -1, 0], # Vertical
[0, bl -1], [bs -1, bl - 1]]
cornersH = [[0, 0], [bl -1, 0], # Horizontal
[0, bs -1], [bl -1, bs - 1]]
# Two adjacent vertical rectangles 4x8 - 4x8 centered on the pixel
blockVL = block(imgE, shift(cornersV, -bs, -bl/2))
blockVR = block(imgE, shift(cornersV, 0, -bl/2))
op1 = let("VL", blockVL,
"VR", blockVR,
IF(GT("VL", "VR"),
THEN(div("VR", "VL")),
ELSE(div("VL", "VR"))))
#op1 = sub(blockVL, blockVR)
#op2 = sub(blockVR, blockVL)
# Two adjacent horizontal rectangles 8x4 - 8x4 centered on the pixel
blockHT = block(imgE, shift(cornersH, -bs, -bl/2))
blockHB = block(imgE, shift(cornersH, -bs, 0))
op3 = let("HT", blockHT,
"HB", blockHB,
IF(GT("HT", "HB"),
THEN(div("HB", "HT")), # div works better than sub
ELSE(div("HT", "HB"))))
#op3 = sub(blockHT, blockHB)
#op4 = sub(blockHB, blockHT)
# Two bright-black-bright vertical features 4x8 - 4x8 - 4x8
block3VL = block(imgE, shift(cornersV, -bs -bs/2, -bl/2))
block3VC = block(imgE, shift(cornersV, -bs/2, -bl/2))
block3VR = block(imgE, shift(cornersV, bs/2, -bl/2))
op5 = let("3VL", block3VL,
"3VC", block3VC,
"3VR", block3VR,
IF(AND(GT("3VL", "3VC"),
GT("3VR", "3VC")),
THEN(sub(add("3VL", "3VR"), "3VC")), # like Viola and Jones 2001
ELSE(div(add("3VL", "3VC", "3VR"), 3)))) # average block value
# Purely like Viola and Jones 2001: work poorly for EM membranes
#op5 = sub(block3VC, block3VL, block3VR) # center minus sides
#op6 = sub(add(block3VL, block3VR), block3VC) # sides minus center
# Two bright-black-bright horizontal features 8x4 / 8x4 / 8x4
block3HT = block(imgE, shift(cornersH, -bl/2, -bs -bs/2))
block3HC = block(imgE, shift(cornersH, -bl/2, -bs/2))
block3HB = block(imgE, shift(cornersH, -bl/2, bs/2))
op7 = let("3HT", block3HT,
"3HC", block3HC,
"3HB", block3HB,
IF(AND(GT("3HT", "3HC"),
GT("3HB", "3HC")),
THEN(sub(add("3HT", "3HB"), "3HC")), # like Viola and Jones 2001
ELSE(div(add("3HT", "3HC", "3HB"), 3)))) # average block value
# Purely like Viola and Jones 2001: work poorly for EM membranes
#op7 = sub(block3HC, block3HT, block3HB) # center minus top and bottom
#op8 = sub(add(block3HT, block3HB), block3HC) # top and bottom minus center
# Combination of vertical and horizontal edge detection
#op9 = maximum(op1, op3)
#op10 = maximum(op6, op8)
#op10 = maximum(op5, op7)
"""
# corners of a square block where 0,0 is at the top left
cornersS = [[0, 0], [bs, 0],
[0, bs], [bs, bs]]
# 2x2 squares for oblique edge detection
blockSTL = block(imgE, shift(cornersS, -bs, -bs)) # top left
blockSTR = block(imgE, shift(cornersS, 0, -bs)) # top right
blockSBL = block(imgE, shift(cornersS, -bs, 0)) # bottom left
blockSBR = block(imgE, cornersS) # bottom right
op11 = sub(add(blockSTL, blockSBR), blockSTR, blockSBL)
op12 = sub(add(blockSTR, blockSBL), blockSTL, blockSBR)
"""
# Combination of vertical, horizontal and oblique edge detection
#op13 = maximum(op1, op3, op6, op8, op11, op12)
#op13 = maximum(op1, op3, op5, op7, op11, op12) # combinations are terrible for RandomForest
# Edge detectors: sum of 3 adjacent pixels (not dividing by the other 6
# to avoid penalizing Y membrane configurations)
"""
op14 = maximum(add(offset(op13, [-1, -1]), op13, offset(op13, [ 1, 1])),
add(offset(op13, [ 0, -1]), op13, offset(op13, [ 0, 1])),
add(offset(op13, [ 1, -1]), op13, offset(op13, [-1, 1])),
add(offset(op13, [-1, 0]), op13, offset(op13, [ 1, 0])))
"""
#opMean, opVariance = filterBankBlockStatistics(img, integralImgE=imgE, sumType=sumType, converter=converter)
# Return an ordered list of all ops
#return [op1, op2, op3, op4, op5, op6, op7, op8, op9, op10, op11, op12, op13, op14]
#return [op1, op3, op5, op7, opMean, opVariance]
return [op1, op3, op5, op7]
def filterBankOrthogonalEdges(img,
bs=4,
bl=8,
sumType=UnsignedLongType(),
converter=Util.genericRealTypeConverter()):
""" Haar-like features from Viola and Jones using integral images of a set of rotated images
tuned to identify neuron membranes in electron microscopy.
bs: length of the short side of a block
bl: length of the long side of a block
sumType: the type with which to add up pixel values in the integral image
converter: for the IntegralImg, to convert from input to sumType
"""
# Create the integral image, stored as 64-bit
alg = IntegralImg(img, sumType, converter)
alg.process()
integralImg = alg.getResult()
imgE = Views.extendBorder(integralImg)
# corners of a 4x8 or 8x4 rectangular block where 0,0 is the top left
cornersV = [[0, 0], [bs -1, 0], # Vertical
[0, bl -1], [bs -1, bl - 1]]
cornersH = [[0, 0], [bl -1, 0], # Horizontal
[0, bs -1], [bl -1, bs - 1]]
# Two adjacent vertical rectangles 4x8 - 4x8 centered on the pixel
blockVL = block(imgE, shift(cornersV, -bs, -bl/2))
blockVR = block(imgE, shift(cornersV, 0, -bl/2))
op1 = sub(blockVL, blockVR)
op2 = sub(blockVR, blockVL)
# Two adjacent horizontal rectangles 8x4 - 8x4 centered on the pixel
blockHT = block(imgE, shift(cornersH, -bs, -bl/2))
blockHB = block(imgE, shift(cornersH, -bs, 0))
op3 = sub(blockHT, blockHB)
op4 = sub(blockHB, blockHT)
# Two bright-black-bright vertical features 4x8 - 4x8 - 4x8
block3VL = block(imgE, shift(cornersV, -bs -bs/2, -bl/2))
block3VC = block(imgE, shift(cornersV, -bs/2, -bl/2))
block3VR = block(imgE, shift(cornersV, bs/2, -bl/2))
op5 = let("3VL", block3VL,
"3VC", block3VC,
"3VR", block3VR,
IF(AND(GT("3VL", "3VC"),
GT("3VR", "3VC")),
THEN(sub(add("3VL", "3VR"), "3VC")), # like Viola and Jones 2001
ELSE(div(add("3VL", "3VC", "3VR"), 3)))) # average block value
# Purely like Viola and Jones 2001: work poorly for EM membranes
#op5 = sub(block3VC, block3VL, block3VR) # center minus sides
#op6 = sub(add(block3VL, block3VR), block3VC) # sides minus center
# Two bright-black-bright horizontal features 8x4 / 8x4 / 8x4
block3HT = block(imgE, shift(cornersH, -bl/2, -bs -bs/2))
block3HC = block(imgE, shift(cornersH, -bl/2, -bs/2))
block3HB = block(imgE, shift(cornersH, -bl/2, bs/2))
op7 = let("3HT", block3HT,
"3HC", block3HC,
"3HB", block3HB,
IF(AND(GT("3HT", "3HC"),
GT("3HB", "3HC")),
THEN(sub(add("3HT", "3HB"), "3HC")),
ELSE(div(add("3HT", "3HC", "3HB"), 3)))) # average block value TODO make a single block
#op7 = sub(block3HC, block3HT, block3HB) # center minus top and bottom
#op8 = sub(add(block3HT, block3HB), block3HC) # top and bottom minus center
# Two adjacent horizontal rectanges, 12x12 - 4x4
bll = bl + bl/2
cornersLarge = [[0, 0], [bll -1, 0],
[0, bll -1], [bll -1, bll -1]]
cornersSmall = [[0, 0], [bs -1, 0],
[0, bs -1], [bs -1, bs -1]]
# Subtract a large black rectangle from a small bright one - aiming at capturing synapses
# Bright on the right
blockLargeL = block(imgE, shift(cornersLarge, -bll, -bll/2))
blockSmallR = block(imgE, shift(cornersSmall, 0, -bs/2))
op9 = sub(blockSmallR, blockLargeL)
# Bright on the left
blockLargeR = block(imgE, shift(cornersLarge, 0, -bll/2))
blockSmallL = block(imgE, shift(cornersSmall, -bs, -bs/2))
op10 = sub(blockSmallL, blockLargeR)
# Bright at the bottom
blockLargeT = block(imgE, shift(cornersLarge, -bll/2, -bll))
blockSmallB = block(imgE, shift(cornersSmall, -bs/2, 0))
op11 = sub(blockSmallB, blockLargeT)
# Bright at the top
blockLargeB = block(imgE, shift(cornersLarge, -bll/2, 0))
blockSmallT = block(imgE, shift(cornersSmall, -bs/2, -bs))
op12 = sub(blockSmallT, blockLargeB)
return [op1, op2, op3, op4, op5, op7, op9, op10, op11, op12]
from net.imglib2.algorithm.math.ImgMath import computeIntoFloat, sub, div from net.imglib2.img.display.imagej import ImageJFunctions as IL from net.imglib2.type.numeric.real import FloatType from ij import IJ # Simple example: normalize an image imp = IJ.getImage() ip = imp.getProcessor() minV, maxV = ip.getMin(), ip.getMax() img = IL.wrap(imp) op = div(sub(img, minV), maxV - minV + 1) result = computeIntoFloat(op) IL.wrap(result, "normalized").show() # As a view IL.wrap(op.view(FloatType()), "normalized").show()
def filterBank(img,
sumType=UnsignedLongType(),
converter=Util.genericRealTypeConverter()):
""" Haar-like features from Viola and Jones
tuned to identify neuron membranes in electron microscopy. """
# Create the integral image, stored as 64-bit
alg = IntegralImg(img, sumType, converter)
alg.process()
integralImg = alg.getResult()
imgE = Views.extendBorder(integralImg)
# corners of a 4x8 or 8x4 rectangular block where 0,0 is the top left
bs = 4 # short side
bl = 8 # long side
cornersV = [
[0, 0],
[bs - 1, 0], # Vertical
[0, bl - 1],
[bs - 1, bl - 1]
]
cornersH = [
[0, 0],
[bl - 1, 0], # Horizontal
[0, bs - 1],
[bl - 1, bs - 1]
]
# Two adjacent vertical rectangles 4x8 - 4x8 centered on the pixel
blockVL = block(imgE, shift(cornersV, -bs, -bl / 2))
blockVR = block(imgE, shift(cornersV, 0, -bl / 2))
op1 = sub(blockVL, blockVR)
op2 = sub(blockVR, blockVL)
# Two adjacent horizontal rectangles 8x4 - 8x4 centered on the pixel
blockHT = block(imgE, shift(cornersH, -bs, -bl / 2))
blockHB = block(imgE, shift(cornersH, -bs, 0))
op3 = sub(blockHT, blockHB)
op4 = sub(blockHB, blockHT)
# Two bright-black-bright vertical features 4x8 - 4x8 - 4x8
block3VL = block(imgE, shift(cornersV, -bs - bs / 2, -bl / 2))
block3VC = block(imgE, shift(cornersV, -bs / 2, -bl / 2))
block3VR = block(imgE, shift(cornersV, bs / 2, -bl / 2))
op5 = sub(block3VC, block3VL, block3VR) # center minus sides
op6 = sub(add(block3VL, block3VR), block3VC) # sides minus center
# Two bright-black-bright horizontal features 8x4 / 8x4 / 8x4
block3HT = block(imgE, shift(cornersH, -bl / 2, -bs - bs / 2))
block3HC = block(imgE, shift(cornersH, -bl / 2, -bs / 2))
block3HB = block(imgE, shift(cornersH, -bl / 2, bs / 2))
op7 = sub(block3HC, block3HT, block3HB) # center minus top and bottom
op8 = sub(add(block3HT, block3HB), block3HC) # top and bottom minus center
# Combination of vertical and horizontal edge detection
op9 = maximum(op1, op3)
op10 = maximum(op6, op8)
# corners of a square block where 0,0 is at the top left
cornersS = [[0, 0], [bs, 0], [0, bs], [bs, bs]]
# 2x2 squares for oblique edge detection
blockSTL = block(imgE, shift(cornersS, -bs, -bs)) # top left
blockSTR = block(imgE, shift(cornersS, 0, -bs)) # top right
blockSBL = block(imgE, shift(cornersS, -bs, 0)) # bottom left
blockSBR = block(imgE, cornersS) # bottom right
op11 = sub(add(blockSTL, blockSBR), blockSTR, blockSBL)
op12 = sub(add(blockSTR, blockSBL), blockSTL, blockSBR)
# Combination of vertical, horizontal and oblique edge detection
op13 = maximum(op1, op3, op6, op8, op11, op12)
# Edge detectors: sum of 3 adjacent pixels (not dividing by the other 6
# to avoid penalizing Y membrane configurations)
op14 = maximum(add(offset(op13, [-1, -1]), op13, offset(op13, [1, 1])),
add(offset(op13, [0, -1]), op13, offset(op13, [0, 1])),
add(offset(op13, [1, -1]), op13, offset(op13, [-1, 1])),
add(offset(op13, [-1, 0]), op13, offset(op13, [1, 0])))
# Return a list of all ops
#return [ob for name, ob in vars().iteritems() if re.match(r"^op\d+$", name)]
# Ordered
return [
op1, op2, op3, op4, op5, op6, op7, op8, op9, op10, op11, op12, op13,
op14
]
def test(red, green, blue, easy=True):
saturation = let(
"red", red, "green", green, "blue", blue, "max",
maximum("red", "green", "blue"), "min",
minimum("red", "green", "blue"),
IF(EQ(0, "max"), THEN(0), ELSE(div(sub("max", "min"), "max"))))
brightness = div(maximum(red, green, blue), 255.0)
hue = IF(
EQ(0, saturation), THEN(0),
ELSE(
let(
"red", red, "green", green, "blue", blue, "max",
maximum("red", "green", "blue"), "min",
minimum("red", "green", "blue"), "range", sub("max", "min"),
"redc", div(sub("max", "red"), "range"), "greenc",
div(sub("max", "green"), "range"), "bluec",
div(sub("max", "blue"), "range"), "hue",
div(
IF(
EQ("red", "max"), THEN(sub("bluec", "greenc")),
ELSE(
IF(EQ("green", "max"),
THEN(sub(add(2, "redc"), "bluec")),
ELSE(sub(add(4, "greenc"), "redc"))))), 6),
IF(LT("hue", 0), THEN(add("hue", 1)), ELSE("hue")))))
#print hierarchy(hue)
#print "hue view:", hue.view( FloatType() ).iterationOrder()
if easy:
# About 26 ms
"""
hsb = Views.stack( hue.view( FloatType() ),
saturation.view( FloatType() ),
brightness.view( FloatType() ) )
"""
# About 13 ms: half! Still much worse than plain ImageJ,
# but the source images are iterated 4 times, rather than just once,
# and the saturation is computed twice,
# and the min, max is computed 3 and 4 times, respectively.
hsb = Views.stack(hue.viewDouble(FloatType()),
saturation.viewDouble(FloatType()),
brightness.viewDouble(FloatType()))
"""
# Even worse: ~37 ms
width, height = rgb.dimension(0), rgb.dimension(1)
h = compute(hue).into(ArrayImgs.floats([width, height]))
s = compute(saturation).into(ArrayImgs.floats([width, height]))
b = compute(brightness).into(ArrayImgs.floats([width, height]))
hsb = Views.stack( h, s, b )
"""
imp = IL.wrap(hsb, "HSB view")
else:
# Tested it: takes more time (~40 ms vs 26 ms above)
width, height = rgb.dimension(0), rgb.dimension(1)
hb = zeros(width * height, 'f')
sb = zeros(width * height, 'f')
bb = zeros(width * height, 'f')
h = ArrayImgs.floats(hb, [width, height])
s = ArrayImgs.floats(sb, [width, height])
b = ArrayImgs.floats(bb, [width, height])
#print "ArrayImg:", b.iterationOrder()
ImgUtil.copy(ImgView.wrap(hue.view(FloatType()), None), h)
ImgUtil.copy(ImgView.wrap(saturation.view(FloatType()), None), s)
ImgUtil.copy(ImgView.wrap(brightness.view(FloatType()), None), b)
stack = ImageStack(width, height)
stack.addSlice(FloatProcessor(width, height, hb, None))
stack.addSlice(FloatProcessor(width, height, sb, None))
stack.addSlice(FloatProcessor(width, height, bb, None))
imp = ImagePlus("hsb", stack)
return imp