def poreDetectionUV(inputImp, inputDataset, inputRoi, ops, data, display, detectionParameters):
title = inputImp.getTitle()
title=title.replace('UV', 'SD')
print title
#trueColorImp= WindowManager.getImage(title)
#print type( trueColorImp)
# calculate are of roi
stats=inputImp.getStatistics()
inputRoiArea=stats.area
print inputRoi
# get the bounding box of the active roi
inputRec = inputRoi.getBounds()
x1=long(inputRec.getX())
y1=long(inputRec.getY())
x2=x1+long(inputRec.getWidth())-1
y2=y1+long(inputRec.getHeight())-1
print x1
print y1
print x2
print y2
# crop the roi
interval=FinalInterval( array([x1, y1 ,0], 'l'), array([x2, y2, 2], 'l') )
cropped=ops.crop(interval, None, inputDataset.getImgPlus() )
datacropped=data.create(cropped)
display.createDisplay("cropped", datacropped)
croppedPlus=IJ.getImage()
duplicator=Duplicator()
substackMaker=SubstackMaker()
# duplicate the roi
duplicate=duplicator.run(croppedPlus)
#duplicate.show()
# convert duplicate of roi to HSB and get brightness
IJ.run(duplicate, "HSB Stack", "");
brightnessPlus=substackMaker.makeSubstack(duplicate, "3-3")
brightness=ImgPlus(ImageJFunctions.wrapByte(brightnessPlus))
brightnessPlus.setTitle("Brightness")
#brightnessPlus.show()
# make another duplicate, split channels and get red
duplicate=duplicator.run(croppedPlus)
channels=ChannelSplitter().split(duplicate)
redPlus=channels[0]
red=ImgPlus(ImageJFunctions.wrapByte(redPlus))
redPlus.show()
# convert to lab
IJ.run(croppedPlus, "Color Transformer", "colour=Lab")
IJ.selectWindow('Lab')
labPlus=IJ.getImage()
# get the A channel
APlus=substackMaker.makeSubstack(labPlus, "2-2")
APlus.setTitle('A')
APlus.show()
APlus.getProcessor().resetMinAndMax()
APlus.updateAndDraw()
AThresholded=threshold(APlus, -10, 50)
# get the B channel
BPlus=substackMaker.makeSubstack(labPlus, "3-3")
BPlus.setTitle('B')
BPlus.show()
BPlus.getProcessor().resetMinAndMax()
BPlus.updateAndDraw()
BThresholded=threshold(BPlus, -10, 50)
# AND the Athreshold and Bthreshold to get a map of the red pixels
ic = ImageCalculator();
redMask = ic.run("AND create", AThresholded, BThresholded);
IJ.run(redMask, "Divide...", "value=255");
#redMask.show()
labPlus.close()
# threshold the spots from the red channel
thresholdedred=SpotDetectionGray(red, data, display, ops, False)
display.createDisplay("thresholdedred", data.create(thresholdedred))
impthresholdedred = ImageJFunctions.wrap(thresholdedred, "wrapped")
# threshold the spots from the brightness channel
thresholded=SpotDetectionGray(brightness, data, display, ops, False)
display.createDisplay("thresholded", data.create(thresholded))
impthresholded=ImageJFunctions.wrap(thresholded, "wrapped")
# or the thresholding results from red and brightness channel
impthresholded = ic.run("OR create", impthresholded, impthresholdedred);
# convert to mask
Prefs.blackBackground = True
IJ.run(impthresholded, "Convert to Mask", "")
# clear the region outside the roi
clone=inputRoi.clone()
clone.setLocation(0,0)
Utility.clearOutsideRoi(impthresholded, clone)
# create a hidden roi manager
roim = RoiManager(True)
# count the particlesimp.getProcessor().setColor(Color.green)
countParticles(impthresholded, roim, detectionParameters.minSize, detectionParameters.maxSize, detectionParameters.minCircularity, detectionParameters.maxCircularity)
# define a function to determine the percentage of pixels that are foreground in a binary image
# inputs:
# imp: binary image, 0=background, 1=foreground
# roi: an roi
def isRed(imp, roi):
stats = imp.getStatistics()
if (stats.mean>detectionParameters.redPercentage): return True
else: return False
def notRed(imp, roi):
stats = imp.getStatistics()
if (stats.mean>detectionParameters.redPercentage): return False
else: return True
allList=[]
for roi in roim.getRoisAsArray():
allList.append(roi.clone())
# count particles that are red
redList=CountParticles.filterParticlesWithFunction(redMask, allList, isRed)
# count particles that are red
blueList=CountParticles.filterParticlesWithFunction(redMask, allList, notRed)
print "Total particles: "+str(len(allList))
print "Filtered particles: "+str(len(redList))
# for each roi add the offset such that the roi is positioned in the correct location for the
# original image
[roi.setLocation(roi.getXBase()+x1, roi.getYBase()+y1) for roi in allList]
# create an overlay and add the rois
overlay1=Overlay()
inputRoi.setStrokeColor(Color.green)
overlay1.add(inputRoi)
[CountParticles.addParticleToOverlay(roi, overlay1, Color.red) for roi in redList]
[CountParticles.addParticleToOverlay(roi, overlay1, Color.cyan) for roi in blueList]
def drawAllRoisOnImage(imp, mainRoi, redList, blueList):
imp.getProcessor().setColor(Color.green)
IJ.run(imp, "Line Width...", "line=3");
imp.getProcessor().draw(inputRoi)
imp.updateAndDraw()
IJ.run(imp, "Line Width...", "line=1");
[CountParticles.drawParticleOnImage(imp, roi, Color.magenta) for roi in redList]
[CountParticles.drawParticleOnImage(imp, roi, Color.green) for roi in blueList]
imp.updateAndDraw()
drawAllRoisOnImage(inputImp, inputRoi, redList, blueList)
#drawAllRoisOnImage(trueColorImp, inputRoi, redList, blueList)
# draw overlay
#inputImp.setOverlay(overlay1)
#inputImp.updateAndDraw()
statsdict=CountParticles.calculateParticleStats(APlus, BPlus, redMask, roim.getRoisAsArray())
print inputRoiArea
areas=statsdict['Areas']
poreArea=0
for area in areas:
poreArea=poreArea+area
ATotal=0
ALevels=statsdict['ALevel']
for A in ALevels:
ATotal=ATotal+A
AAverage=ATotal/len(ALevels)
BTotal=0
BLevels=statsdict['BLevel']
for B in BLevels:
BTotal=BTotal+B
BAverage=BTotal/len(BLevels)
redTotal=0
redPercentages=statsdict['redPercentage']
for red in redPercentages:
redTotal=redTotal+red
redAverage=redTotal/len(redPercentages)
pixwidth=inputImp.getCalibration().pixelWidth
inputRoiArea=inputRoiArea/(pixwidth*pixwidth)
print str(len(allList))+" "+str(len(redList))+" "+str(len(blueList))+" "+str(poreArea/inputRoiArea)+" "+str(redAverage)
