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

[FJGEODEV]

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

[FJGEODEV]

Hi, just check other issues reported last year and found it is same issue, but I still can;t figure it out, here is some output:

print(delD_norm.min(),delD_norm.max())
0.0 380.9160012391183

with np.errstate(divide='ignore',invalid='ignore'):
delD_xn = delD_x / delD_norm
delD_yn = delD_y / delD_norm
print(delD_xn.min(),delD_xn.max())
print(delD_yn.min(),delD_yn.max())

nan nan
nan nan

print(mir_delD_xn.shape)
print(mir_delD_yn.shape)
(1603, 1335)
(1603, 1335)

Could you just try my image? The white color is 1, black is 0.

Thanks.