def test_principal_boxes(self):
self.image.reshape_for_landmarks(2)
self.image.threshold_image()
threshold = self.image.get_uint_image()
self.image.skeletonization()
skeleton = self.image.get_uint_image()
self.image.bin_to_bgr()
skeleton_rgb = self.image.get_uint_image()
landmarks, segments = l.potential_landmarks(skeleton, 3)
widths = l.vessel_width(threshold, landmarks)
vessels = l.finding_landmark_vessels(widths, landmarks, skeleton,
skeleton_rgb)
marked_skeleton, final_landmarks = l.vessel_number(
vessels, landmarks, skeleton_rgb)
bifurcations, crossings = l.principal_boxes(marked_skeleton,
final_landmarks, 2)
result = np.genfromtxt(self._test_path + "boxes_bifurcations_test.csv",
delimiter=',')
result2 = np.genfromtxt(self._test_path + "boxes_crossings_test.csv",
delimiter=',')
assert_array_equal(result, bifurcations[0],
"Bifurcation points does not match")
assert_array_equal(result2, crossings[0],
"Crossing points does not match")
def test_vessel_number(self):
self.image.reshape_for_landmarks(2)
self.image.threshold_image()
threshold = self.image.get_uint_image()
self.image.skeletonization()
skeleton = self.image.get_uint_image()
self.image.bin_to_bgr()
skeleton_rgb = self.image.get_uint_image()
landmarks, segments = l.potential_landmarks(skeleton, 3)
widths = l.vessel_width(threshold, landmarks)
vessels = l.finding_landmark_vessels(widths, landmarks, skeleton,
skeleton_rgb)
marked_skeleton, final_landmarks = l.vessel_number(
vessels, landmarks, skeleton_rgb)
result_skeleton = np.genfromtxt(self._test_path +
"vessel_number_skeleton_test.csv",
delimiter=',')
result_landmarks = np.genfromtxt(self._test_path +
"vessel_number_filter_test.csv",
delimiter=',')
assert_array_equal(result_skeleton, marked_skeleton[20, :],
"Vessel skeleton does not match")
assert_array_equal(result_landmarks, final_landmarks,
"Vessel final landmarks does not match")
def _feature_vectors():
directory = _base_directory_training + "original/"
features = []
for filename in sorted(glob.glob(os.path.join(directory, '*.tif'))):
name = os.path.basename(filename)
name = name.split(".")[0]
original = cv2.imread(filename, 1)
gray = cv2.imread(filename, 0)
(minVal, maxVal, minLoc, maxLoc) = cv2.minMaxLoc(gray)
lab = cv2.cvtColor(original, cv2.COLOR_BGR2LAB)
L, A, B = cv2.split(lab)
manual = retina.Retina(
None, _base_directory_training + "manual/" + name + ".png")
manual.threshold_image()
thr_img = manual.get_uint_image()
cv2.circle(thr_img, maxLoc, 60, 0, -1)
manual.skeletonization()
skeleton_img = manual.get_uint_image()
cv2.circle(skeleton_img, maxLoc, 60, 0, -1)
landmarks, segmented_skeleton_img = l.potential_landmarks(
skeleton_img, 3)
av = cv2.imread(_base_directory_training + "av/" + name + ".png", 1)
widths = _vessel_widths(segmented_skeleton_img, thr_img)
data = _preparing_data(widths, 6, original, av, L, gray)
features.extend(data)
h5f = h5py.File(_base_directory_model + 'vector_features_interpolation.h5',
'w')
h5f.create_dataset('training', data=features)
return features
def test_vessel_width(self):
self.manual.threshold_image()
threshold = self.manual.get_uint_image()
self.manual.skeletonization()
skeleton = self.manual.get_uint_image()
landmarks, segmented = l.potential_landmarks(skeleton, 3)
widths = vc._vessel_widths(segmented, threshold)
result = np.genfromtxt(self._test_path + "vessels_width_test.csv", delimiter=',')
assert_array_equal(result, widths[0:10], "Vessel widths does not match")
def test_potential_landmarks(self):
self.image.reshape_for_landmarks(2)
self.image.threshold_image()
self.image.skeletonization()
skeleton = self.image.get_uint_image()
landmarks, segments = l.potential_landmarks(skeleton, 3)
result = np.genfromtxt(self._test_path +
"potential_landmarks_test.csv",
delimiter=',')
assert_array_equal(result, landmarks, "landmark points does not match")
def test_vessel_width(self):
self.image.reshape_for_landmarks(2)
self.image.threshold_image()
threshold = self.image.get_uint_image()
self.image.skeletonization()
skeleton = self.image.get_uint_image()
landmarks, segments = l.potential_landmarks(skeleton, 3)
widths = l.vessel_width(threshold, landmarks)
result = np.genfromtxt(self._test_path + "vessel_widths_test.csv",
delimiter=',')
assert_array_equal(result, widths, "Vessel widths does not match")
def test_preparing_data_without_av(self):
self.manual.threshold_image()
threshold = self.manual.get_uint_image()
self.manual.skeletonization()
skeleton = self.manual.get_uint_image()
gray = cv2.cvtColor(self.original, cv2.COLOR_BGR2GRAY)
lab = cv2.cvtColor(self.original, cv2.COLOR_BGR2LAB)
L, A, B = cv2.split(lab)
landmarks, segmented = l.potential_landmarks(skeleton, 3)
widths = vc._vessel_widths(segmented, threshold)
features = vc._preparing_data(widths, 6, self.original, None, L, gray)
result = np.genfromtxt(self._test_path + "preparing_data_without_av_test.csv", delimiter=',')
assert_array_equal(result, features[0:10], "Data does not match")
def test_finding_landmark_vessels(self):
self.image.reshape_for_landmarks(2)
self.image.threshold_image()
threshold = self.image.get_uint_image()
self.image.skeletonization()
skeleton = self.image.get_uint_image()
self.image.bin_to_bgr()
skeleton_rgb = self.image.get_uint_image()
landmarks, segments = l.potential_landmarks(skeleton, 3)
widths = l.vessel_width(threshold, landmarks)
vessels = l.finding_landmark_vessels(widths, landmarks, skeleton,
skeleton_rgb)
result = np.genfromtxt(self._test_path +
"finding_landmark_vessels_test.csv",
delimiter=',')
assert_array_equal(result, vessels[0],
"Landmark vessels does not match")
def _loading_model(original: np.ndarray, threshold: np.ndarray, av: np.ndarray,
size: int):
# Load model of the neuronal network
json_file = open(_base_directory_model + 'modelVA.json', "r")
loaded_model_json = json_file.read()
json_file.close()
loaded_model = model_from_json(loaded_model_json)
# Load weights
loaded_model.load_weights(_base_directory_model + 'modelVA.h5')
gray = cv2.cvtColor(original, cv2.COLOR_BGR2GRAY)
(minVal, maxVal, minLoc, maxLoc) = cv2.minMaxLoc(gray)
lab = cv2.cvtColor(original, cv2.COLOR_BGR2LAB)
L, A, B = cv2.split(lab)
manual = retina.Retina(threshold, None)
manual.threshold_image()
thr_img = manual.get_uint_image()
cv2.circle(thr_img, maxLoc, 60, 0, -1)
manual.skeletonization()
skeleton_img = manual.get_uint_image()
cv2.circle(skeleton_img, maxLoc, 60, 0, -1)
landmarks, segmented_skeleton_img = l.potential_landmarks(skeleton_img, 3)
widths = _vessel_widths(segmented_skeleton_img, thr_img)
data = _preparing_data(widths, 6, original, av, L, gray)
features = np.array(data)
predict_img = np.full(
(segmented_skeleton_img.shape[0], segmented_skeleton_img.shape[1]),
3,
dtype=float)
for row in range(0, features.shape[0]):
prediction = loaded_model.predict(np.divide(
features[row:row + 1, 2:size], 255),
batch_size=1)
predict_img[features[row, 0], features[row, 1]] = prediction
return features, segmented_skeleton_img, thr_img, predict_img
