def test_ray_features_circle_down_edge(self):
seg = np.zeros((400, 600), dtype=bool)
x, y = draw.circle(200, 250, 150, shape=seg.shape)
seg[x, y] = True
points = [(200, 250), (150, 200), (250, 200), (250, 300)]
for i, point in enumerate(points):
ray_dist_raw = seg_fts.compute_ray_features_segm_2d(
seg, point, angle_step=ANGULAR_STEP, edge='down')
ray_dist, shift = seg_fts.shift_ray_features(ray_dist_raw)
points = seg_fts.reconstruct_ray_features_2d(
point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points, ray_dist_raw,
ray_dist, 'circle_e-down-A-%i.png' % i)
self.assertTrue(os.path.exists(p_fig))
x, y = draw.circle(200, 250, 120, shape=seg.shape)
seg[x, y] = False
for i, point in enumerate(points):
ray_dist_raw = seg_fts.compute_ray_features_segm_2d(
seg, point, angle_step=ANGULAR_STEP, edge='down')
ray_dist, shift = seg_fts.shift_ray_features(ray_dist_raw)
points = seg_fts.reconstruct_ray_features_2d(
point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points, ray_dist_raw,
ray_dist, 'circle_e-down-B-%i.png' % i)
self.assertTrue(os.path.exists(p_fig))
def test_ray_features_circle_down_edge(self):
seg = np.zeros((400, 600), dtype=bool)
x, y = draw.circle(200, 250, 150, shape=seg.shape)
seg[x, y] = True
points = [(200, 250), (150, 200), (250, 200), (250, 300)]
for i, point in enumerate(points):
ray_dist_raw = compute_ray_features_segm_2d(
seg, point, edge='down', angle_step=ANGULAR_STEP)
ray_dist, shift = shift_ray_features(ray_dist_raw)
points_rt = reconstruct_ray_features_2d(point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points_rt, ray_dist_raw,
ray_dist,
'circle-full_edge-down-%i.png' % i)
self.assertTrue(os.path.isfile(p_fig))
# insert white interior
x, y = draw.circle(200, 250, 120, shape=seg.shape)
seg[x, y] = False
for i, point in enumerate(points):
ray_dist_raw = compute_ray_features_segm_2d(
seg, point, edge='down', angle_step=ANGULAR_STEP)
ray_dist, shift = shift_ray_features(ray_dist_raw)
points_rt = reconstruct_ray_features_2d(point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points_rt, ray_dist_raw,
ray_dist,
'circle-inter_edge-down-%i.png' % i)
self.assertTrue(os.path.isfile(p_fig))
def prepare_boundary_points_ray_join(seg,
centers,
close_points=5,
min_diam=MIN_ELLIPSE_DAIM,
sel_bg=STRUC_ELEM_BG,
sel_fg=STRUC_ELEM_FG):
""" extract some point around foreground boundaries
:param ndarray seg: input segmentation
:param [(int, int)] centers: list of centers
:param float close_points: remove closest point then a given threshold
:param int min_diam: minimal size of expected objest
:param int sel_bg: smoothing background with morphological operation
:param int sel_fg: smoothing foreground with morphological operation
:return [ndarray]:
>>> seg = np.zeros((10, 20), dtype=int)
>>> ell_params = 5, 10, 4, 6, np.deg2rad(30)
>>> seg = add_overlap_ellipse(seg, ell_params, 1)
>>> pts = prepare_boundary_points_ray_join(seg, [(4, 9)], 5, 3,
... sel_bg=1, sel_fg=0)
>>> np.round(pts).tolist() # doctest: +NORMALIZE_WHITESPACE
[[[4.0, 16.0],
[7.0, 10.0],
[9.0, 6.0],
[1.0, 9.0],
[4.0, 16.0],
[7.0, 10.0],
[1.0, 9.0]]]
"""
seg_bg, seg_fg = split_segm_background_foreground(seg, sel_bg, sel_fg)
points_centers = []
for center in centers:
ray_bg = seg_fts.compute_ray_features_segm_2d(seg_bg, center)
ray_bg[ray_bg < min_diam] = min_diam
points_bg = seg_fts.reconstruct_ray_features_2d(center, ray_bg)
points_bg = seg_fts.reduce_close_points(points_bg, close_points)
ray_fc = seg_fts.compute_ray_features_segm_2d(seg_fg,
center,
edge='down')
ray_fc[ray_fc < min_diam] = min_diam
points_fc = seg_fts.reconstruct_ray_features_2d(center, ray_fc)
points_fc = seg_fts.reduce_close_points(points_fc, close_points)
points_both = np.vstack((points_bg, points_fc))
points_centers.append(points_both)
return points_centers
def prepare_boundary_points_ray_mean(
seg,
centers,
close_points=5,
min_diam=MIN_ELLIPSE_DAIM,
sel_bg=STRUC_ELEM_BG,
sel_fg=STRUC_ELEM_FG,
):
""" extract some point around foreground boundaries
:param ndarray seg: input segmentation
:param [(int, int)] centers: list of centers
:param float close_points: remove closest point then a given threshold
:param int min_diam: minimal size of expected objest
:param int sel_bg: smoothing background with morphological operation
:param int sel_fg: smoothing foreground with morphological operation
:return [ndarray]:
>>> seg = np.zeros((10, 20), dtype=int)
>>> ell_params = 5, 10, 4, 6, np.deg2rad(30)
>>> seg = add_overlap_ellipse(seg, ell_params, 1)
>>> pts = prepare_boundary_points_ray_mean(seg, [(4, 9)], 2.5, 3, sel_bg=1, sel_fg=0)
>>> np.round(pts).tolist() # doctest: +NORMALIZE_WHITESPACE
[[[4.0, 16.0],
[7.0, 15.0],
[9.0, 5.0],
[4.0, 5.0],
[1.0, 7.0],
[0.0, 14.0]]]
"""
seg_bg, seg_fc = split_segm_background_foreground(seg, sel_bg, sel_fg)
points_centers = []
for center in centers:
ray_bg = compute_ray_features_segm_2d(seg_bg, center)
ray_fc = compute_ray_features_segm_2d(seg_fc, center, edge='down')
# replace not found (-1) by large values
rays = np.array([ray_bg, ray_fc], dtype=float)
rays[rays < 0] = np.inf
rays[rays < min_diam] = min_diam
# take the smalles from both
ray_min = np.min(rays, axis=0)
ray_mean = np.mean(rays, axis=0)
ray_mean[np.isinf(ray_mean)] = ray_min[np.isinf(ray_mean)]
points_close = reconstruct_ray_features_2d(center, ray_mean)
points_close = reduce_close_points(points_close, close_points)
points_centers.append(points_close)
return points_centers
def test_ray_features_circle(self):
seg = np.ones((400, 600), dtype=bool)
x, y = draw.circle(200, 250, 100, shape=seg.shape)
seg[x, y] = False
points = [(200, 250), (150, 200), (250, 200), (250, 300)]
for i, point in enumerate(points):
ray_dist_raw = compute_ray_features_segm_2d(
seg, point, angle_step=ANGULAR_STEP)
ray_dist, shift = shift_ray_features(ray_dist_raw)
points = reconstruct_ray_features_2d(point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points, ray_dist_raw,
ray_dist, 'circle-%i.png' % i)
self.assertTrue(os.path.isfile(p_fig))
def prepare_boundary_points_ray_dist(seg,
centers,
close_points=1,
sel_bg=STRUC_ELEM_BG,
sel_fg=STRUC_ELEM_FG):
""" extract some point around foreground boundaries
:param ndarray seg: input segmentation
:param [(int, int)] centers: list of centers
:param float close_points: remove closest point then a given threshold
:param int sel_bg: smoothing background with morphological operation
:param int sel_fg: smoothing foreground with morphological operation
:return [ndarray]:
>>> seg = np.zeros((10, 20), dtype=int)
>>> ell_params = 5, 10, 4, 6, np.deg2rad(30)
>>> seg = add_overlap_ellipse(seg, ell_params, 1)
>>> pts = prepare_boundary_points_ray_dist(seg, [(4, 9)], 2, sel_bg=0, sel_fg=0)
>>> np.round(pts, 2).tolist() # doctest: +NORMALIZE_WHITESPACE
[[[4.0, 16.0],
[6.8, 15.0],
[9.0, 5.5],
[4.35, 5.0],
[1.0, 6.9],
[1.0, 9.26],
[0.0, 11.31],
[0.5, 14.0],
[1.45, 16.0]]]
"""
seg_bg, _ = split_segm_background_foreground(seg, sel_bg, sel_fg)
points = []
for center in centers:
ray = compute_ray_features_segm_2d(seg_bg, center)
points_bg = reconstruct_ray_features_2d(center, ray, 0)
points_bg = reduce_close_points(points_bg, close_points)
points += points_bg.tolist()
points = np.array(points)
# remove all very small negative valeue, probaly by rounding
points[(points < 0) & (points > -1e-3)] = 0.
dists = spatial.distance.cdist(points, centers, metric='euclidean')
close_center = np.argmin(dists, axis=1)
points_centers = []
for i in range(close_center.max() + 1):
pts = points[close_center == i]
points_centers.append(pts)
return points_centers
def test_ray_features_ellipse(self):
seg = np.ones((400, 600), dtype=bool)
x, y = draw.ellipse(200, 250, 120, 200, rotation=np.deg2rad(30),
shape=seg.shape)
seg[x, y] = False
points = [(200, 250), (150, 200), (250, 300)]
for i, point in enumerate(points):
ray_dist_raw = seg_fts.compute_ray_features_segm_2d(
seg, point, angle_step=ANGULAR_STEP)
# ray_dist, shift = seg_fts.shift_ray_features(ray_dist_raw)
points = seg_fts.reconstruct_ray_features_2d(point, ray_dist_raw)
p_fig = export_ray_results(seg, point, points, ray_dist_raw, [],
'ellipse-%i.png' % i)
self.assertTrue(os.path.exists(p_fig))
def test_ray_features_polygon(self):
seg = np.ones((400, 600), dtype=bool)
x, y = draw.polygon(np.array([50, 170, 300, 250, 150, 150, 50]),
np.array([100, 270, 240, 150, 150, 80, 50]),
shape=seg.shape)
seg[x, y] = False
centres = [(150, 200), (200, 250), (250, 200), (120, 100)]
for i, point in enumerate(centres):
ray_dist_raw = compute_ray_features_segm_2d(
seg, point, angle_step=ANGULAR_STEP)
ray_dist, shift = shift_ray_features(ray_dist_raw)
points = reconstruct_ray_features_2d(point, ray_dist, shift)
p_fig = export_ray_results(seg, point, points, ray_dist_raw,
ray_dist, 'polygon-%i.png' % i)
self.assertTrue(os.path.isfile(p_fig))
def prepare_boundary_points_ray_dist(seg,
centers,
close_points=1,
sel_bg=STRUC_ELEM_BG,
sel_fg=STRUC_ELEM_FG):
""" extract some point around foreground boundaries
:param ndarray seg: input segmentation
:param [(int, int)] centers: list of centers
:param float close_points: remove closest point then a given threshold
:param int sel_bg: smoothing background with morphological operation
:param int sel_fg: smoothing foreground with morphological operation
:return [ndarray]:
>>> seg = np.zeros((10, 20), dtype=int)
>>> ell_params = 5, 10, 4, 6, np.deg2rad(30)
>>> seg = add_overlap_ellipse(seg, ell_params, 1)
>>> pts = prepare_boundary_points_ray_dist(seg, [(4, 9)], 2,
... sel_bg=0, sel_fg=0)
>>> np.round(pts).tolist() # doctest: +NORMALIZE_WHITESPACE
[[[0.0, 2.0],
[4.0, 16.0],
[6.0, 15.0],
[9.0, 6.0],
[6.0, 5.0],
[3.0, 7.0],
[0.0, 10.0]]]
"""
seg_bg, _ = split_segm_background_foreground(seg, sel_bg, sel_fg)
points = np.array((0, np.asarray(centers).shape[1]))
for center in centers:
ray = seg_fts.compute_ray_features_segm_2d(seg_bg, center)
points_bg = seg_fts.reconstruct_ray_features_2d(center, ray, 0)
points_bg = seg_fts.reduce_close_points(points_bg, close_points)
points = np.vstack((points, points_bg))
dists = spatial.distance.cdist(points, centers, metric='euclidean')
close_center = np.argmin(dists, axis=1)
points_centers = []
for i in range(close_center.max() + 1):
pts = points[close_center == i]
points_centers.append(pts)
return points_centers