def segment_veins(img, test=False):
to_plot = []
""" Segmentamos venas """
img_inverted = skimage.util.invert(img)
if test:
to_plot.append(("inverted green channel", img_inverted))
img_inverted = exposure.equalize_adapthist(img_inverted, clip_limit=0.5)
if test:
to_plot.append(("high contrast", img_inverted))
mask_invert = th.mask_otsu(img_inverted)
if test:
to_plot.append(("Global Otsu", th.apply(img, mask_invert)))
""" creamos mascara para eliminar fondo """
(ancho, alto) = img.shape
rr, cc = ellipse(ancho / 2, alto / 2, (ancho / 2) - 5, (alto / 2) - 5)
mask_background = np.zeros((ancho, alto)) > 0
mask_background[rr, cc] = 1
mask_invert = np.logical_and(mask_invert, mask_background)
mask_invert = skimage.morphology.remove_small_objects(mask_invert, 200)
if test:
to_plot.append(("Mask without background", mask_invert))
if test:
vi.plot_multy(to_plot, 1, 4, 'Veins segmentation')
return mask_invert, mask_background
def clean(input_file, thresh_val = [250, 245, 240, 230, 225, 220],
window_size = 5, kernel_size = 5):
# Ensures that both quality and window_size parameters are integers
window_size = int(window_size)
kernel_size = int(kernel_size)
#Checking arguments and raising expected exceptions
check_arguments(input_file, window_size, kernel_size)
# Loading the image
image = input_file
if isinstance(image, str):
image = cv.imread(input_file)
# Applying Grayscale, Gaussian and median blur and erode
image = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
image = cv.GaussianBlur(image, (window_size, window_size), 0)
image = cv.medianBlur(image, window_size)
image = cv.erode(image, (kernel_size, kernel_size))
# Applying threshold list
results = th.apply(image, thresh_val)
return results
def closest_prop(img, mask):
(lbls, num) = label(mask, return_num=True)
max_mean = 0.0
max_mask = None
if num > 1:
for lb in range(num + 1):
current_mean = 0
current = th.apply(img, lbls == lb)
current_mean = np.sum(current) / np.count_nonzero(current)
if max_mean < current_mean:
max_mean = current_mean
max_mask = current
return max_mask > 0
else:
return mask
def clean(input_file, thresh_val = [225, 220, 215, 210, 205, 200],
window_size = 3):
# Ensures that window_size parameter is integer
window_size = int(window_size)
# Checking arguments and raising expected exceptions
check_arguments(window_size)
# Loading the image
image = iu.get_image(input_file)
# Gray-scale and Gaussian Blur
image = cv.cvtColor(image, cv.COLOR_BGR2GRAY)
image = cv.GaussianBlur(image, (window_size, window_size), 0)
# Applying threshold list
results = th.apply(image, thresh_val)
return results
def handmade_p_tile_method(img, op="disc", test=False):
if op == "disc" or op == "cup":
to_plot = []
img_red = img[:, :, 0]
if test:
to_plot.append(("Red Channel", img_red))
img_red = dilation(img_red, disk(3))
if test:
to_plot.append(("Gaussing", img_red))
img_red = enhance_contrast(img_red, disk(10))
if test:
to_plot.append(("Enhace_contrast", img_red))
mask = f.hand_made(img_red, 10, 2)
if test:
to_plot.append(("P-tile", th.apply(img_red, mask)))
mask = binary_opening(mask, disk(7))
mask = skimage.morphology.remove_small_objects(mask, 1700)
if test:
to_plot.append(
("Opening and remove small objects", th.apply(img_red, mask)))
props = msr.props(img_red, mask)[0]
minr, minc, maxr, maxc = props.bbox
img_cut = img[minr:maxr, minc:maxc]
if test:
to_plot.append(("Binary opening", th.apply(img_red, mask)))
if test:
vi.plot_multy(to_plot, 2, 3, 'P-tile')
if op == "cup":
minr, minc, maxr, maxc = props.bbox
img_green = img[minr:maxr, minc:maxc, 1]
to_plot = []
if test:
to_plot.append(("green channel", img_green))
v_mask, _ = segment_veins(img_green, test)
img_aux = closing(img_green, disk(6))
img_aux = dilation(img_aux, disk(6))
img_aux2 = dilation(img_green, disk(3))
if test:
to_plot.append(("closed + dilated", img_aux))
to_plot.append(("dilated", img_aux2))
img_v_closed = th.apply(img_aux, v_mask)
img_t_dilated = th.apply(img_aux2, v_mask == False)
if test:
to_plot.append(("veins part", img_v_closed))
to_plot.append(("target part", img_t_dilated))
img_green = img_v_closed + img_t_dilated
if test:
to_plot.append(("without veins", img_green))
img_green = dilation(img_green, disk(6))
img_green = enhance_contrast(img_green, disk(10))
if test:
to_plot.append(("dilation + contrast", img_green))
mask = f.hand_made(img_green, 10, 2, smallest=0)
if test:
to_plot.append(("P-tile", th.apply(img_green, mask)))
mask1 = mask
mask = np.zeros(img[:, :, 1].shape)
mask[minr:maxr, minc:maxc] = mask1
if test:
vi.plot_multy(to_plot, 4, 4, 'Otsu Local chain')
return (mask, img_cut, props)
def otsu_local_method(img, op="disc", test=False):
if op == "disc" or op == "cup":
to_plot = []
img_red = img[:, :, 0]
img_red = skimage.util.img_as_ubyte(img_red)
if test:
to_plot.append(("red_chan", img_red))
img_red = skimage.filters.gaussian(img_red, 4.2)
if test:
to_plot.append(("gaussian f", img_red))
img_red = enhance_contrast(img_red, disk(3))
if test:
to_plot.append(("enhace_contrast", img_red))
""" Local Otsu """
t_loc_otsu = otsu(img_red, disk(20))
mask = img_red >= t_loc_otsu
if test:
to_plot.append(("Otsu local", th.apply(img_red, mask)))
mask = binary_opening(mask, disk(2))
mask = binary_closing(mask, disk(4))
if test:
to_plot.append(("binary open-close", th.apply(img_red, mask)))
""" Chusing brightest region """
mask = msr.closest_prop(img_red, mask)
if test:
to_plot.append(("brightest region", th.apply(img_red, mask)))
maskfinal = binary_closing(mask, disk(20))
if test:
to_plot.append(("binary closing", th.apply(img_red, maskfinal)))
img_red = th.apply(img_red, maskfinal)
props = msr.props(img_red, maskfinal)[0]
minr, minc, maxr, maxc = props.bbox
img_cut = img[minr:maxr, minc:maxc]
if test:
vi.plot_multy(to_plot, 3, 3, 'Otsu Local chain')
if op == "cup":
minr, minc, maxr, maxc = props.bbox
img_green = img[minr:maxr, minc:maxc, 1]
to_plot = []
if test:
to_plot.append(("green channel", img_green))
v_mask, _ = segment_veins(img_green, test)
img_aux = closing(img_green, disk(6))
img_aux = dilation(img_aux, disk(6))
img_aux2 = dilation(img_green, disk(3))
if test:
to_plot.append(("closed + dilated", img_aux))
to_plot.append(("dilated", img_aux2))
img_v_closed = th.apply(img_aux, v_mask)
img_t_dilated = th.apply(img_aux2, v_mask == False)
if test:
to_plot.append(("veins part", img_v_closed))
to_plot.append(("target part", img_t_dilated))
img_green = img_v_closed + img_t_dilated
if test:
to_plot.append(("img_green", img_green))
""" Local Otsu """
t_loc_otsu = otsu(img_green, disk(20))
mask = img_green >= t_loc_otsu
mask = msr.closest_prop(img_green, mask)
if test:
to_plot.append(("Otsu local", th.apply(img_green, mask)))
mask1 = mask
mask = np.zeros(img[:, :, 1].shape)
mask[minr:maxr, minc:maxc] = mask1
if test:
vi.plot_multy(to_plot, 2, 4, 'cup')
return (mask, img_cut, props)
def sobel_watershed_method(img, op="disc", veins=False, test=False):
if op == "disc" or op == "cup":
to_plot = []
img_red = img[:, :, 0]
if test:
to_plot.append(("Red Channel", img_red))
img_red = skimage.util.img_as_ubyte(img_red)
img_red = skimage.filters.gaussian(img_red, 0.1)
if test:
to_plot.append(("Gaussian Filter", img_red))
img_red = enhance_contrast(img_red, disk(6))
if test:
to_plot.append(("Enhace Contrast", img_red))
elevation_map = sobel(img_red)
if test:
to_plot.append(("gradientes", elevation_map))
markers = np.zeros_like(img_red)
s2 = f.target_set_mean(img_red, 8.5)
markers[img_red < 150] = 1
markers[img_red > s2] = 2
seg_img = segmentation.watershed(elevation_map, markers)
mask = (seg_img - 1) > 0
if test:
to_plot.append(("Sobel + WaterShed", seg_img))
mask = binary_opening(mask, disk(2))
mask = skimage.morphology.remove_small_objects(mask, 400)
if test:
to_plot.append(("Removing small objects", th.apply(img_red, mask)))
mask = binary_closing(mask, disk(6))
if test:
to_plot.append(("Closing Region", th.apply(img[:, :, 0], mask)))
mask = skimage.morphology.remove_small_objects(mask, 1700)
if test:
to_plot.append(
("Removing Big Objects", th.apply(img[:, :, 0], mask)))
mask = binary_closing(mask, disk(6))
mask = msr.closest_prop(img[:, :, 0], mask)
if test:
to_plot.append(
("Removing non brighter region", th.apply(img[:, :, 0], mask)))
mask = binary_dilation(mask, disk(3))
mask = binary_closing(mask, disk(12))
if test:
to_plot.append(
("Dilate result region", th.apply(img[:, :, 0], mask)))
props = msr.props(img_red, mask)[0]
minr, minc, maxr, maxc = props.bbox
img_cut = img[minr:maxr, minc:maxc]
if test:
vi.plot_multy(to_plot, 3, 4, 'Sobel Watershed')
if op == "cup":
minr, minc, maxr, maxc = props.bbox
img_green = img[minr:maxr, minc:maxc, 1]
to_plot = []
columns = 1
if test:
to_plot.append(("green channel", img_green))
if not veins:
columns = 4
v_mask, _ = segment_veins(img_green, test)
img_aux = closing(img_green, disk(6))
img_aux = dilation(img_aux, disk(6))
img_aux2 = dilation(img_green, disk(3))
if test:
to_plot.append(("closed + dilated", img_aux))
to_plot.append(("dilated", img_aux2))
img_v_closed = th.apply(img_aux, v_mask)
img_t_dilated = th.apply(img_aux2, v_mask == False)
if test:
to_plot.append(("veins part", img_v_closed))
to_plot.append(("target part", img_t_dilated))
img_green = img_v_closed + img_t_dilated
if test:
to_plot.append(("without veins", img_green))
img_green = dilation(img_green, disk(6))
img_green = enhance_contrast(img_green, disk(10))
elevation_map = sobel(img_green)
markers = np.zeros_like(img_green)
s2 = f.target_set_mean(img_green, 8.5)
markers[img_green < 150] = 1
markers[img_green > s2] = 2
seg_img = segmentation.watershed(elevation_map, markers)
mask = (seg_img - 1) > 0
if test:
to_plot.append(("P-tile", th.apply(img_green, mask)))
mask1 = mask
mask = np.zeros(img[:, :, 1].shape)
mask[minr:maxr, minc:maxc] = mask1
if test:
vi.plot_multy(to_plot, 2, columns, 'cup')
return (mask, img_cut, props)