mmkamani7 / skeletonization Goto Github PK

i tried the code with my image (512x512 ) but it did not work , what can i change to make it work for my data ?

I run your code I got error no module named fuzzyTransform

Duplicates of #2 #5.
File "Skeletonization.py", line ..., in
fluxMap = flux(mir_delD_xn, mir_delD_yn)
File "skeleton2Graph.py", line 195, in flux
flux_x = Nx * delD_xn[pix[0] - 1:pix[0] + 2, pix[1] - 1:pix[1] + 2]
ValueError: operands could not be broadcast together with shapes (3,3) (0,0)
According to your instructions in #2, Igot
mir_delD_yn.shape: (284, 284)
the outputs of this line to be correct and non empty:
D,IDX = morph.distance_transform_edt(BW,return_distances=True, return_indices=True)

Help needed.

Hello, thanks for your code. When I run the demo, I got the problem saying "IndexError: only integers, slices (:), ellipsis (...), numpy.newaxis (None) and integer or boolean arrays are valid indices" skelenton2Graph.py, line 329, in skeleton2Graph edgeLinkedLength = np.array(edgeLength)[map(int, list(np.abs(edgeLinkedNumber) - 1))].tolist(). My work environment is python3.x, windows. Maybe this is the different version of python that map function return different type result, can you help me solve this problem?

Hi, thanks for this good work, I was trying your method on my own image.

When calculating flux, I had this error message:

ValueError Traceback (most recent call last)
in
1 #Calculate flux map
----> 2 fluxMap = flux(mir_delD_xn,mir_delD_yn)
3 plt.imshow(np.nan_to_num(fluxMap))
4 plt.title('Flux Map')
5 plt.show()

~/CODES/Skeletonization-master/skeleton2Graph.py in flux(delD_xn, delD_yn)
199 print(pix[0],pix[1])
200 print(delD_xn[pix[0]-1:pix[0]+2,pix[1]-1:pix[1]+2].shape)
--> 201 flux_x = Nx * delD_xn[pix[0]-1:pix[0]+2,pix[1]-1:pix[1]+2]
202 flux_y = Ny * delD_yn[pix[0]-1:pix[0]+2,pix[1]-1:pix[1]+2]
203 flux_x[1,1] = np.nan

ValueError: operands could not be broadcast together with shapes (3,3) (0,0)

I print out some more information in flux function:

Nx = -1/np.sqrt(2) * np.array([[-1, 0, 1],[-np.sqrt(2), 0, np.sqrt(2)],[-1, 0, 1]])
print('NX=', Nx.shape)
Ny = -1/np.sqrt(2) * np.array([[-1, -np.sqrt(2), -1],[0, 0, 0],[1, np.sqrt(2), 1]])
print('NY=', Ny.shape)
flux = np.zeros(delD_xn.shape)
print('flux=', flux.shape)

flux.fill(np.nan)
nonNanPix = np.argwhere(np.invert(np.isnan(delD_xn) | np.isnan(delD_yn)))
print('nonNanPix=', nonNanPix.shape)

for pix in nonNanPix:
print(pix[0],pix[1])
print(delD_xn[pix[0]-1:pix[0]+2,pix[1]-1:pix[1]+2].shape)
..............................

NX= (3, 3)
NY= (3, 3)
flux= (1603, 1335)
nonNanPix= (1456285, 2)
0 0
(0, 0)

It looks that something wrong to find out nonNanPix, but I don;t understand why, I also attached my image, could you take a look? Thanks.

Fan