Files
scikit-image/skimage/feature/censure.py
T

284 lines
9.6 KiB
Python

import numpy as np
from scipy.ndimage.filters import maximum_filter, minimum_filter, convolve
from ..transform import integral_image
from ..feature.corner import _compute_auto_correlation
from ..util import img_as_float
from ..morphology import convex_hull_image
from .censure_cy import _censure_dob_loop, _slanted_integral_image, _censure_octagon_loop
def _get_filtered_image(image, n_scales, mode):
# TODO : Implement the STAR mode
scales = np.zeros((image.shape[0], image.shape[1], n_scales), dtype=np.double)
if mode == 'DoB':
for i in range(n_scales):
n = i + 1
# Constant multipliers for the outer region and the inner region
# of the bilevel filters with the constraint of keeping the DC bias
# 0.
inner_weight = (1.0 / (2 * n + 1)**2)
outer_weight = (1.0 / (12 * n**2 + 4 * n))
integral_img = integral_image(image)
filtered_image = np.zeros(image.shape)
_censure_dob_loop(image, n, integral_img, filtered_image, inner_weight, outer_weight)
scales[:, :, i] = filtered_image
elif mode == 'Octagon':
# TODO : Decide the shapes of Octagon filters for scales > 7
outer_shape = [(5, 2), (5, 3), (7, 3), (9, 4), (9, 7), (13, 7), (15, 10)]
inner_shape = [(3, 0), (3, 1), (3, 2), (5, 2), (5, 3), (5, 4), (5, 5)]
for i in range(n_scales):
scales[:, :, i] = convolve(image, _octagon_filter(outer_shape[i][0], outer_shape[i][1], inner_shape[i][0], inner_shape[i][1]))
"""
integral_img = integral_image(image)
integral_img1 = _slanted_integral_image_modes(image, 1)
integral_img2 = _slanted_integral_image_modes(image, 2)
integral_img3 = _slanted_integral_image_modes(image, 3)
integral_img4 = _slanted_integral_image_modes(image, 4)
for k in range(n_scales):
n = k + 1
filtered_image = np.zeros(image.shape)
mo = outer_shape[n - 1][0]
no = outer_shape[n - 1][1]
mi = inner_shape[n - 1][0]
ni = inner_shape[n - 1][1]
outer_pixels = (mo + 2 * no)**2 - 2 * no * (no + 1)
inner_pixels = (mi + 2 * ni)**2 - 2 * ni * (ni + 1)
outer_weight = 1.0 / (outer_pixels - inner_pixels)
inner_weight = 1.0 / inner_pixels
_censure_octagon_loop(image, integral_img, integral_img1, integral_img2, integral_img3, integral_img4, filtered_image, outer_weight, inner_weight, mo, no, mi, ni)
scales[:, :, k] = filtered_image
"""
return scales
def _oct(m, n):
f = np.zeros((m + 2*n, m + 2*n))
f[0, n] = 1
f[n, 0] = 1
f[0, m + n -1] = 1
f[m + n - 1, 0] = 1
f[-1, n] = 1
f[n, -1] = 1
f[-1, m + n - 1] = 1
f[m + n - 1, -1] = 1
return convex_hull_image(f).astype(int)
def _octagon_filter(mo, no, mi, ni):
outer = (mo + 2 * no)**2 - 2 * no * (no + 1)
inner = (mi + 2 * ni)**2 - 2 * ni * (ni + 1)
outer_wt = 1.0 / (outer - inner)
inner_wt = 1.0 / inner
c = ((mo + 2 * no) - (mi + 2 * ni)) / 2
outer_oct = _oct(mo, no)
inner_oct = np.zeros((mo + 2 * no, mo + 2 * no))
inner_oct[c:-c, c:-c] = _oct(mi, ni)
bfilter = outer_wt * outer_oct - (outer_wt + inner_wt) * inner_oct
return bfilter
def _filter_using_convolve(image, n, mode='DoB'):
if mode == 'DoB':
inner_wt = (1.0 / (2*n + 1)**2)
outer_wt = (1.0 / (12*n**2 + 4*n))
dob_filter = np.zeros((4 * n + 1, 4 * n + 1))
dob_filter[:] = outer_wt
dob_filter[n : 3 * n + 1, n : 3 * n + 1] = - inner_wt
return convolve(image, dob_filter)
elif mode == 'Octagon':
outer_shape = [(5, 2), (5, 3), (7, 3), (9, 4), (9, 7), (13, 7), (15, 10)]
inner_shape = [(3, 0), (3, 1), (3, 2), (5, 2), (5, 3), (5, 4), (5, 5)]
return convolve(image, _octagon_filter(outer_shape[n - 1][0], outer_shape[n - 1][1], inner_shape[n - 1][0], inner_shape[n - 1][1]))
def _slanted_integral_image_modes(img, mode=1):
if mode == 1:
"""
The following figures describe area that is summed up to calculate
the value at point @ in slanted integral image. The subtended at @ is
135 degrees.
censure_cy._slanted_integral_image performs the mode1
_slanted_integral_image
_________________
|********/ |
|*******/ |
|******/ |
|-----@ |
| |
| |
|_________________|
"""
image = np.copy(img, order='C')
mode1 = np.zeros((image.shape[0] + 1, image.shape[1]), order='C')
_slanted_integral_image(image, mode1)
return mode1[1:, :]
elif mode == 2:
"""
For mode2, the image can be first flipped left-right and then up-down.
Then we can use censure_cy._slanted_integral_image and the returned
result can be flipped left-right and then up-down to get the following
mode.
_________________
| |
| |
| |
| @_____|
| /******|
| /*******|
|________/________|
"""
image = np.copy(img, order='C')
image = np.fliplr(image)
image = np.flipud(image)
mode2 = np.zeros((image.shape[0] + 1, image.shape[1]), order='C')
_slanted_integral_image(image, mode2)
mode2 = mode2[1:, :]
mode2 = np.fliplr(mode2)
mode2 = np.flipud(mode2)
return mode2
elif mode == 3:
"""
_________________
| |
|\\ |
|*\\ |
|**\\ |
|***@ |
|***| |
|___|_____________|
"""
image = np.copy(img, order='C')
image = np.flipud(image)
image = image.T
mode3 = np.zeros((image.shape[0] + 1, image.shape[1]), order='C')
_slanted_integral_image(image, mode3)
mode3 = mode3[1:, :]
mode3 = np.flipud(mode3.T)
return mode3
else:
"""
________________
| |****|
| |****|
| @****|
| \\**|
| \\*|
| \\|
|________________|
"""
image = np.copy(img, order='C')
image = np.fliplr(image)
image = image.T
mode4 = np.zeros((image.shape[0] + 1, image.shape[1]), order='C')
_slanted_integral_image(image, mode4)
mode4 = mode4[1:, :]
mode4 = np.fliplr(mode4.T)
return mode4
def _suppress_line(response, sigma, rpc_threshold):
Axx, Axy, Ayy = _compute_auto_correlation(response, sigma)
detA = Axx * Ayy - Axy**2
traceA = Axx + Ayy
# ratio of principal curvatures
rpc = traceA**2 / (detA + 0.001)
response[rpc > rpc_threshold] = 0
return response
def censure_keypoints(image, n_scales=7, mode='DoB', threshold=0.03, rpc_threshold=10):
"""
Extracts Censure keypoints along with the corresponding scale using
either Difference of Boxes, Octagon or STAR bilevel filter.
Parameters
----------
image : 2D ndarray
Input image.
n_scales : positive integer
Number of scales to extract keypoints from. The keypoints will be
extracted from all the scales except the first and the last.
mode : {'DoB', 'Octagon', 'STAR'}
Type of bilevel filter used to get the scales of input image. Possible
values are 'DoB', 'Octagon' and 'STAR'.
threshold : float
Threshold value used to suppress maximas and minimas with a weak
magnitude response obtained after Non-Maximal Suppression.
rpc_threshold : float
Threshold for rejecting interest points which have ratio of principal
curvatures greater than this value.
Returns
-------
keypoints : (N, 3) array
Location of extracted keypoints along with the corresponding scale.
References
----------
.. [1] Motilal Agrawal, Kurt Konolige and Morten Rufus Blas
"CenSurE: Center Surround Extremas for Realtime Feature
Detection and Matching",
http://link.springer.com/content/pdf/10.1007%2F978-3-540-88693-8_8.pdf
.. [2] Adam Schmidt, Marek Kraft, Michal Fularz and Zuzanna Domagala
"Comparative Assessment of Point Feature Detectors and
Descriptors in the Context of Robot Navigation"
http://www.jamris.org/01_2013/saveas.php?QUEST=JAMRIS_No01_2013_P_11-20.pdf
"""
image = np.squeeze(image)
if image.ndim != 2:
raise ValueError("Only 2-D gray-scale images supported.")
image = img_as_float(image)
image = np.ascontiguousarray(image)
# Generating all the scales
scales = _get_filtered_image(image, n_scales, mode)
# Suppressing points that are neither minima or maxima in their 3 x 3 x 3
# neighbourhood to zero
minimas = (minimum_filter(scales, (3, 3, 3)) == scales) * scales
maximas = (maximum_filter(scales, (3, 3, 3)) == scales) * scales
# Suppressing minimas and maximas weaker than threshold
minimas[np.abs(minimas) < threshold] = 0
maximas[np.abs(maximas) < threshold] = 0
response = maximas + minimas
for i in range(1, n_scales - 1):
# sigma = (window_size - 1) / 6.0
# window_size = 7 + 2 * i
# Hence sigma = 1 + i / 3.0
response[:, :, i] = _suppress_line(response[:, :, i], (1 + i / 3.0), rpc_threshold)
# Returning keypoints with its scale
keypoints = np.transpose(np.nonzero(response[:, :, 1:n_scales - 1])) + [0, 0, 2]
return keypoints