def test_threshold_adaptive_gaussian(self):
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[ True, True, False, False, False]]
)
out = threshold_adaptive(self.image, 3, method='gaussian')
assert_equal(ref, out)
out = threshold_adaptive(self.image, 3, method='gaussian', param=1.0 / 3.0)
assert_equal(ref, out)
def test_threshold_adaptive_gaussian(self):
ref = np.array([[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]])
out = threshold_adaptive(self.image, 3, method='gaussian')
assert_equal(ref, out)
out = threshold_adaptive(self.image,
3,
method='gaussian',
param=1. / 3.)
assert_equal(ref, out)
def test_threshold_adaptive_median(self):
ref = np.array([[False, False, False, False, True],
[False, False, True, False, False],
[False, False, True, False, False],
[False, False, True, True, False],
[False, True, False, False, False]])
out = threshold_adaptive(self.image, 3, method='median')
assert_equal(ref, out)
def test_threshold_adaptive_median(self):
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, False],
[False, False, True, False, False],
[False, False, True, True, False],
[False, True, False, False, False]]
)
out = threshold_adaptive(self.image, 3, method='median')
assert_equal(ref, out)
def test_threshold_local_equals_adaptive(self):
def func(arr):
return arr.sum() / arr.shape[0]
with expected_warnings(['deprecated', 'return value']):
thresholded_original = threshold_adaptive(self.image, 3,
method='generic',
param=func)
threshold_new = threshold_local(self.image, 3, method='generic',
param=func)
assert_equal(thresholded_original, self.image > threshold_new)
def test_threshold_local_equals_adaptive(self):
def func(arr):
return arr.sum() / arr.shape[0]
with expected_warnings(['deprecated', 'return value']):
thresholded_original = threshold_adaptive(self.image, 3,
method='generic',
param=func)
threshold_new = threshold_local(self.image, 3, method='generic',
param=func)
assert_equal(thresholded_original, self.image > threshold_new)
def test_threshold_adaptive_generic(self):
def func(arr):
return arr.sum() / arr.shape[0]
ref = np.array([[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[True, True, False, False, False]])
out = threshold_adaptive(self.image, 3, method='generic', param=func)
assert_equal(ref, out)
def test_threshold_adaptive_generic(self):
def func(arr):
return arr.sum() / arr.shape[0]
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[ True, True, False, False, False]]
)
out = threshold_adaptive(self.image, 3, method='generic', param=func)
assert_equal(ref, out)
def test_threshold_adaptive_generic(self):
def func(arr):
return arr.sum() / arr.shape[0]
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[ True, True, False, False, False]]
)
with expected_warnings(['deprecated', 'return value']):
out = threshold_adaptive(self.image, 3, method='generic',
param=func)
assert_equal(ref, out)
def test_threshold_adaptive_generic(self):
def func(arr):
return arr.sum() / arr.shape[0]
ref = np.array(
[[False, False, False, False, True],
[False, False, True, False, True],
[False, False, True, True, False],
[False, True, True, False, False],
[ True, True, False, False, False]]
)
with expected_warnings(['deprecated', 'return value']):
out = threshold_adaptive(self.image, 3, method='generic',
param=func)
assert_equal(ref, out)
from skimage.filters.thresholding import threshold_adaptive
from skimage.morphology import square, erosion
from skimage.exposure import rescale_intensity
from skimage.measure import label, regionprops
from scipy.ndimage import binary_fill_holes
import numpy as np
from barcode import draw_regions
if __name__ == '__main__':
img = imread('./../../images/bloodcells.jpg')
p1, p2 = np.percentile(img, (0, 90))
img = rescale_intensity(img, in_range=(p1, p2))
img_gray = rgb2gray(img)
img_gray_filtered = 1. - threshold_adaptive(
img_gray, block_size=31, offset=0.02)
img_gray_filtered = erosion(img_gray_filtered, selem=square(3))
filled = binary_fill_holes(img_gray_filtered)
labels = label(filled)
regions = regionprops(labels)
cells = []
for region in regions:
bbox = region.bbox
height = bbox[2] - bbox[0]
width = bbox[3] - bbox[1]
if 20 < height and 10 < width:
cells.append(region)
print "Number if cells:", len(cells)
cells_result = draw_regions(cells, img_gray.shape)
plt.imshow(cells_result, 'gray')
plt.show()
from matplotlib import pyplot as plt
from skimage.color import rgb2gray
from skimage.filters.thresholding import threshold_adaptive
from skimage.morphology import opening, square
from skimage.measure import label, regionprops
def draw_regions(regs, img_size):
img = np.ndarray((img_size[0], img_size[1]), dtype='float64')
for reg in regs:
coords = reg.coords
for coord in coords:
img[coord[0]][coord[1]] = 1.
return img
if __name__ == '__main__':
barcode_img = imread('./../../images/barcode.jpg')
barcode_gray = rgb2gray(barcode_img)
barcode_filtered = 1. - threshold_adaptive(barcode_gray, block_size=75, offset=0.04)
barcode_filtered = opening(barcode_filtered, selem=square(3))
labels = label(barcode_filtered)
regions = regionprops(labels)
num_regions = []
for region in regions:
bbox = region.bbox
height = bbox[2] - bbox[0]
width = bbox[3] - bbox[1]
if 25 < height < 45 and 1 <= float(height) / width <= 6 and (region.extent < 0.57 or region.extent > 0.80):
num_regions.append(region)
barcode_result = draw_regions(num_regions, barcode_gray.shape)
plt.imshow(barcode_result, 'gray')
plt.show()