def denoising_lowermiginification_guassianblur(inputsvs,magnification,dictx,patch_size ,upperlimit, lowerlimit):
non_black_img,slide_dimen = readWSI(inputsvs,magnification = "5x")
#sliden = np.asarray(non_black_img, dtype="int32")
sliden1 = localization_with_GaussianBlur(inputsvs,non_black_img,patch_size ,upperlimit, lowerlimit)
# print("Slide demensions",slide_dimen)
sliden2 = mask_generation(sliden1,slide_dimen[dictx[magnification]],mask_generation_c = "G")
garbage_collector()
del(sliden1,non_black_img,slide_dimen)
return sliden2
def denoising(inputsvs,
magnification="20x",
filtering="GaussianBlur",
patch_size=(256, 256),
upperlimit=1100,
lowerlimit=300,
red_value=(80, 220),
green_value=(80, 200),
blue_value=(80, 170),
Annotation=None,
Annotatedlevel=0,
Requiredlevel=0):
slide = OpenSlide(inputsvs)
slide_dimensions = slide.level_dimensions
img, slide_dimensions = readWSI(inputsvs, magnification, Annotation,
Annotatedlevel, Requiredlevel)
dictx = dictionary(slide_dimensions)
if filtering == "GaussianBlur":
out = denoising_using_GaussianBlur(inputsvs, magnification, dictx,
patch_size, upperlimit, lowerlimit,
Annotation, Annotatedlevel,
Requiredlevel)
elif filtering == "RGB":
mask = denoising_RGB_Thersholding(img, slide_dimensions, magnification,
dictx, patch_size, red_value,
green_value, blue_value)
out = np.zeros_like(img)
print("cleaning image at high mignification")
mask = mask.astype(np.bool)
out[mask] = img[mask]
out = np.where(out != [0, 0, 0], out, [255, 255, 255])
print("cleaning WSI done")
# cv2.imwrite("/home/pagenet2/PageNet2/Data Preprocessing Pipeline/WSI_Precessing_test/cleanedimages/%s/cleanedsmallf.png"%(inputsvs.split("/")[-1][:-4]),out)
garbage_collector()
print("exisiting cleaning")
else:
mask = denoising_No_filters(img, slide_dimensions, magnification,
dictx)
out = np.zeros_like(img)
print("cleaning image at high mignification")
mask = mask.astype(np.bool)
out[mask] = img[mask]
out = np.where(out != [0, 0, 0], out, [255, 255, 255])
print("cleaning WSI done")
# cv2.imwrite("/home/pagenet2/PageNet2/Data Preprocessing Pipeline/WSI_Precessing_test/cleanedimages/%s/cleanedsmallf.png"%(inputsvs.split("/")[-1][:-4]),out)
garbage_collector()
print("exisiting cleaning")
return out
def localization_with_GaussianBlur(inputsvs,img,patch_size ,upperlimit, lowerlimit):
slide1 = inputsvs
slide1, slide_dims = readWSI(inputsvs)
patch_x = 20
for i in range(int((len(slide1[0]))/patch_x)+1):
for j in range(int((len(slide1))/patch_x)+1):
sample_img = slide1[j*patch_x:j*patch_x+patch_x,i*patch_x:i*patch_x+patch_x]
sample_img_new = GaussianBlur(sample_img,(patch_x,patch_x),upperlimit, lowerlimit)
if sample_img_new is None:
slide1[j * patch_x:j * patch_x + patch_x, i * patch_x:i * patch_x+ patch_x] = np.zeros_like(sample_img)
else:
slide1[j*patch_x:j*patch_x+patch_x,i*patch_x:i*patch_x+patch_x] = sample_img
slide1 = np.where(slide1 != [0, 0, 0], slide1, [255, 255, 255])
garbage_collector()
return slide1
def denoising_lowermiginification_guassianblur(inputsvs,magnification,dictx,patch_size ,upperlimit, lowerlimit, Annotation , Annotatedlevel , Requiredlevel):
non_black_img,slide_dimen = readWSI(inputsvs,"5x",Annotation , Annotatedlevel , Requiredlevel = Requiredlevel+2)
#sliden = np.asarray(non_black_img, dtype="int32")
# magnification = "5x"
sliden1 = localization_with_GaussianBlur(non_black_img,slide_dimen,patch_size ,upperlimit, lowerlimit,Annotation , Annotatedlevel , Requiredlevel)
# print("Slide demensions",slide_dimen)
magnification = magnification
print(magnification)
try:
# cv2.imwrite("check1.png",sliden1)
sliden2 = mask_generation(sliden1,slide_dimen[dictx[magnification]],mask_generation_c = "G")
except:
print("this image has lower magnification then the given magnification, cleaning the image at highest magnification possible")
sliden2 = mask_generation(sliden1,slide_dimen[0],mask_generation_c = "G")
garbage_collector()
del(sliden1,non_black_img,slide_dimen)
return sliden2
def denoising_using_GaussianBlur(inputsvs,magnification,dictx,patch_size ,upperlimit, lowerlimit,Annotation , Annotatedlevel , Requiredlevel):
mask = denoising_lowermiginification_guassianblur(inputsvs,magnification,dictx,patch_size ,upperlimit, lowerlimit,Annotation , Annotatedlevel , Requiredlevel)
# mask = cv2.bitwise_not(mask)
print("cleanedimage at low maginfication done")
img,slide_dimensions= readWSI(inputsvs,magnification ,Annotation , Annotatedlevel , Requiredlevel)
img = mask*img
img[np.where((img == [0,0,0]).all(axis = 2))] = [255,255,255]
# print("loading high magnification image")
# out = np.zeros_like(img)
# img = np.asarray(img, dtype="int32")
# cv2.imwrite("check2.png",img)
# print("cleaning image at high mignification")
# mask = mask.astype(np.bool)
# out[mask] = img[mask]
# cv2.imwrite("check3.png",out)
# out[np.where((out == [0,0,0]).all(axis = 2))] = [255,255,255]
# # out = black_to_white(out)
# print("cleaning WSI done")
# cv2.imwrite("check4.png",out)
garbage_collector()
print("exisiting cleaning")
return img