wassname/scikit-image
1
0
Fork 0
mirror of https://github.com/wassname/scikit-image.git synced 2026-06-30 02:22:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #192 from tangofoxtrotmike/master

Browse Source 
BUG: feature.hog() fails on non-square images.
This commit is contained in:
Stefan van der Walt
2012-05-22 10:54:35 -07:00
parent 8f43ee7739 4d999c313f
commit 582aa193c5
2 changed files with 10 additions and 10 deletions
Download Patch File Download Diff File Expand all files Collapse all files
skimage/feature/hog.py
+8 -8
View File
@@ -97,7 +97,7 @@ def hog(image, orientations=9, pixels_per_cell=(8, 8),
magnitude = sqrt(gx ** 2 + gy ** 2)
orientation = arctan2(gy, (gx + 1e-15)) * (180 / pi) + 90
sx, sy = image.shape
sy, sx = image.shape
cx, cy = pixels_per_cell
bx, by = cells_per_block
@@ -105,7 +105,7 @@ def hog(image, orientations=9, pixels_per_cell=(8, 8),
n_cellsy = int(np.floor(sy // cy)) # number of cells in y
# compute orientations integral images
orientation_histogram = np.zeros((n_cellsx, n_cellsy, orientations))
orientation_histogram = np.zeros((n_cellsy, n_cellsx, orientations))
for i in range(orientations):
#create new integral image for this orientation
# isolate orientations in this range
@@ -118,7 +118,7 @@ def hog(image, orientations=9, pixels_per_cell=(8, 8),
cond2 = temp_ori > 0
temp_mag = np.where(cond2, magnitude, 0)
orientation_histogram[:,:,i] = uniform_filter(temp_mag, size=(cx, cy))[cx/2::cx, cy/2::cy].T
orientation_histogram[:,:,i] = uniform_filter(temp_mag, size=(cy, cx))[cy/2::cy, cx/2::cx]
# now for each cell, compute the histogram
@@ -140,7 +140,7 @@ def hog(image, orientations=9, pixels_per_cell=(8, 8),
dy = radius * sin(float(o) / orientations * np.pi)
rr, cc = draw.bresenham(centre[0] - dx, centre[1] - dy,
centre[0] + dx, centre[1] + dy)
hog_image[rr, cc] += orientation_histogram[x, y, o]
hog_image[rr, cc] += orientation_histogram[y, x, o]
"""
The fourth stage computes normalisation, which takes local groups of
@@ -159,14 +159,14 @@ def hog(image, orientations=9, pixels_per_cell=(8, 8),
n_blocksx = (n_cellsx - bx) + 1
n_blocksy = (n_cellsy - by) + 1
normalised_blocks = np.zeros((n_blocksx, n_blocksy,
bx, by, orientations))
normalised_blocks = np.zeros((n_blocksy, n_blocksx,
by, bx, orientations))
for x in range(n_blocksx):
for y in range(n_blocksy):
block = orientation_histogram[x:x + bx, y:y + by, :]
block = orientation_histogram[y:y + by, x:x + bx, :]
eps = 1e-5
normalised_blocks[x, y, :] = block / sqrt(block.sum() ** 2 + eps)
normalised_blocks[y, x, :] = block / sqrt(block.sum() ** 2 + eps)
"""
The final step collects the HOG descriptors from all blocks of a dense
skimage/feature/tests/test_hog.py
+2 -2
View File
@@ -5,12 +5,12 @@ from skimage.feature import hog
def test_histogram_of_oriented_gradients():
# Replace with skimage.data.lena() after merge
img = scipy.misc.lena().astype(np.int8)
img = scipy.misc.lena()[:256,:].astype(np.int8)
fd = hog(img, orientations=9, pixels_per_cell=(8, 8),
cells_per_block=(1, 1))
assert len(fd) == 9 * (512//8) ** 2
assert len(fd) == 9 * (256//8) * (512//8)
if __name__ == '__main__':
from numpy.testing import run_module_suite