def analizeXMBlobs(listXMarkers):
listXMarkers = inA.resizeAll(listXMarkers)
if len(listXMarkers) != 6:
return 'Problem to find all tests'
listBinaryXMarkers = getXMarkersBinary(listXMarkers)
listXMarkersBlobs = getXMarkersBlobs(listBinaryXMarkers)
control = listXMarkers[5]
if(isinstance(listXMarkersBlobs, str)): #If string received
return listXMarkersBlobs
statsH, statsS, statsV = inA.controlStats(control) #Hue and Saturation control stats
dict_test = dict()
markerNames = ["ESAT6","CFP10","RV1681","P24","H2", "Control"]
for i,img in enumerate(listXMarkersBlobs):
blobs = inA.blobDetect(img)
dict_test[markerNames[i] + "_blobs"] = len(blobs)
dict_test[markerNames[i] + "_area"] = inA.areaEstimation(blobs)
dict_test[markerNames[i] + "_area_color"] = inA.colorSegmentation(listXMarkers[i], statsH, statsS)[2]
#Resultado algoritmo conteo blobs
dict_test[markerNames[i] + "_result_blobs"] = inA.blobAnalysis(listBinaryXMarkers[i], blobs)
'''
#Resultado algoritmo areas de blobs
dict_test[markerNames[i] + "_result_area"] = inA.areaAnalysis(listBinaryXMarkers[i],
dict_test[markerNames[i] + "_area"], i)
'''
#Resultado algoritmo areas de blobs (1 Solo Blob siempre es negativo)
dict_test[markerNames[i] + "_result_area"] = inA.areaAnalysis(listBinaryXMarkers[i], blobs,
dict_test[markerNames[i] + "_area"], i)
dict_test['tb_result_area'] = inA.tbDiagnostic(dict_test[markerNames[0]+ "_result_area"],
dict_test[markerNames[1]+ "_result_area"], dict_test[markerNames[2]+ "_result_area"])
return dict_test
# Checkpoint
#cv2.imwrite(str(i) + 'indRespy.png', cv2.cvtColor(test,cv2.COLOR_RGB2BGR))
plt.show()
if (not batch):
for i, img in enumerate(listKMeansTests):
plt.subplot(321 + i), plt.imshow(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
# Checkpoint
#cv2.imwrite(str(i) + 'indRespy.png', cv2.cvtColor(img,cv2.COLOR_RGB2BGR))
plt.show()
# %%
# Blob detection
listBlobs = []
for i, img in enumerate(listKMeansTestsBina):
imgCopy = img.copy()
blobs = inA.blobDetect(img)
listBlobs.append(blobs)
if (not batch):
if (operator.mod(i, 2) != 0): # Print two by line
print("Blobs detected: " + str(len(listBlobs[i - 1])) +
" Blobs detected: " + str(len(listBlobs[i])))
plt.subplot(321 + i), plt.imshow(inA.drawBlobs(imgCopy, blobs), 'gray')
plt.show()
# Test results using blobs
for i, img in enumerate(listTestsBinary):
if (operator.mod(i, 2) != 0): # Print two by line
if (not batch):
print("Result: " + inA.blobAnalysis(img, listBlobs[i - 1]) +
" Area: " +
str(round(inA.areaEstimation(listBlobs[i - 1]), 2)) +
"\t Result: " + inA.blobAnalysis(img, listBlobs[i]) +
kMeansReconstructedReshaped, 1)
# Grayscale image
imgGray = cv2.cvtColor(kMeansReconstructedReshaped,
cv2.COLOR_BGR2GRAY)
# Thresholded image
_, imgBina = cv2.threshold(imgGray, 160, 255, cv2.THRESH_BINARY)
# And operation with mask
imgBinaMask = inA.andOperation(imgBina, mask)
# Opening of the final image (Erotion followed by dilation)
kernel = np.ones((1, 1), np.uint8)
imgBinaMaskOpen = cv2.morphologyEx(imgBinaMask, cv2.MORPH_OPEN,
kernel)
# Image preprocessing => FINISH
# Blob analysis => START
blobs = inA.blobDetect(imgBinaMaskOpen)
area = round(inA.areaEstimation(blobs), 2)
diagnosis = inA.areaAnalysis2(imgBina, blobs, area,
marker['marker'])
# Blob analysis => FINISH
# Diagnosis comparison => START
kMeansStats = {'diagnosis': diagnosis}
dbStats = {'diagnosis': marker['diagnostic']}
if (kMeansStats['diagnosis'] == dbStats['diagnosis']):
results['coincidences'] += 1
elif (kMeansStats['diagnosis'] != dbStats['diagnosis']):
results['errors'] += 1
# Diagnosis comparison => FINISH
# Image shows (Only one per execution)
if count > 0:
if(isinstance(originalImage, str)):
print(originalImage)
continue
else:
# Image preprocessing => START
imgBina = pP.contourBinarizationOtsu(
originalImage, 3, 3, 45, 4, Gs=0, inverse=False, mean=True)
imgBinMask = inA.andOperation(imgBina, mask)
imgBinMaskMorph = inA.erosionDilation(imgBinMask, 3)
imgBinMaskMorphInv = cv2.bitwise_not(imgBinMaskMorph)
# Image preprocessing => FINISH
# Blob analysis => START
imgBinaMaskInvMorphCopy = imgBinMaskMorphInv.copy()
blobs = inA.blobDetect(imgBinMaskMorphInv)
area = round(inA.areaEstimation(blobs), 2)
diagnosis = inA.areaAnalysis2(
imgBina, blobs, area, marker['marker'])
# Blob analysis => FINISH
# Diagnosis comparison => START
blobsStats = {
'diagnosis': diagnosis
}
dbStats = {
'diagnosis': marker['diagnostic']
}
if (blobsStats['diagnosis'] == dbStats['diagnosis']):
results['coincidences'] += 1
elif (blobsStats['diagnosis'] != dbStats['diagnosis']):