You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello. I'm having trouble segmenting images that are approximately 640x480 in size. In this case, the following problem occurs:
Traceback (most recent call last):
File "./analysis_multi_nC.py", line 155, in <module>
main()
File "./analysis_multi_nC.py", line 119, in main
sp_label = bosupix.label(nC)
File "boruvka_superpixel_wrap.pyx", line 442, in hers_superpixel.BoruvkaSuperpixel.label
raise RuntimeError('illegal n_supix')
RuntimeError: illegal n_supix
When looking at the code, the error occurs as output of the label function in file boruvka_superpixel_wrap.pyx due to the output of function self._bosupix.label(n_supix). The function self._bosupix.label(n_supix), in file boruvka_superpixel.cpp returns NULL when n_supix is less than min_n_supix(). When changing this condition to n_supix equal to min_n_supix(), the image generates some superpixels of size 1.
This problem occurs regardless of the number of superpixels and disappears when resizing the original image. However, I need to process the image to its original size. Any idea how I can modify the code for this purpose?
The text was updated successfully, but these errors were encountered:
Hello. I'm having trouble segmenting images that are approximately 640x480 in size. In this case, the following problem occurs:
When looking at the code, the error occurs as output of the
label
function in fileboruvka_superpixel_wrap.pyx
due to the output of functionself._bosupix.label(n_supix)
. The functionself._bosupix.label(n_supix)
, in fileboruvka_superpixel.cpp
returns NULL whenn_supix
is less thanmin_n_supix()
. When changing this condition ton_supix
equal tomin_n_supix()
, the image generates some superpixels of size 1.This problem occurs regardless of the number of superpixels and disappears when resizing the original image. However, I need to process the image to its original size. Any idea how I can modify the code for this purpose?
The text was updated successfully, but these errors were encountered: