I am wondering how to convert a numpy array in the form [x, y] (positions of pixels):
[[ 93 58]
[ 94 58]
[ 95 58]
...,
[ 99 142]
[100 142]
[101 142]]
to a form that works with skimage. To do this I think I need the array normalized to some canvas size width x height that fits the dataset (say 500 x 500).
Ultimately I want to perform edge/contour detection on this array.
http://scikit-image.org/docs/dev/auto_examples/edges/plot_contours.html
How can I normalize this data so that its in the form skimage requires?
Without calling any external functions, this is a simple way to transform/convert a series of x,y datapoints into a 2 dimensional array suitable for use in skimage:
def xy_to_binary2d(ts):
'''Convert a list of (x,y) tuples to binary 2d format acceptable to skimage.'''
if ts.dtype != 'int32':
print('Only integer input is supported.')
xmax,ymax = ts.max(axis=0)
__,ymin = ts.min(axis=0)
if ymin < 0:
print('Negative integers are not supported.')
r = np.zeros((ymax+2,xmax+2))
for each in ts:r.itemset(each[1],each[0])
return r
Lets test it:
ts =np.array([[1,1],[2,1],[3,2],[4,3],[5,5],[6,8],[7,13]])
xy_to_binary2d(ts)
Output:
array([[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 1., 1., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 1., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 1., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 1., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 1., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 1., 0.],
[ 0., 0., 0., 0., 0., 0., 0., 0., 0.]])
And, a pretty picture... plt.imshow(xy_to_binary2d(ts))
Found a decent solution. Use matplotlib to generate an rgb representation of the pixel locations [x,y] numpy > use skimage color.rgb2gray to convert this to skimage format.
fig = Figure()
canvas = FigureCanvas(fig)
ax = fig.add_subplot(111)
ax.axis('off')
fig.patch.set_facecolor('white')
print(data.shape)
ax.fill(data[:, [0]], data[:, [1]],'black')
# Set canvas size
mi = min(x_min, y_min)
ma = max(x_max, y_max)
ax.set_xlim(mi, ma)
ax.set_ylim(mi, ma)
fig.canvas.draw()
buf = fig.canvas.tostring_rgb()
ncols, nrows = fig.canvas.get_width_height()
new_data = np.fromstring(buf, dtype=np.uint8).reshape(nrows, ncols, 3)
new_data.reshape(nrows, ncols, 3)
image = color.rgb2gray(new_data)
return image
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.