0

I trained the network for my data. Once I ran this: To see the probability map

plt.imshow(prob[1], cmap='hot', interpolation='nearest')
plt.show()

And this is the probability map of Object 1:enter image description here

How can I get the solid segmentation from the probability map (even considering those points with lower probability belonging to the class)?

Thanks

S.EB
  • 1,966
  • 4
  • 29
  • 54
  • 3
    It's hard to give advice without more details of the data, your decision methods/thresholds, etc. Also, asking "how do I do these standard tasks" makes the question "too broad" on two fronts: too many questions in one posting, and lack of research before posting. – Prune Jun 08 '17 at 20:44

1 Answers1

0

There are several ways to get the solid segmentation. One is to use one of the clustering algorithms to define the area of interest. Perhaps a simple threshold cut-off p <= (p > threshold) followed by a density-fill algorithm would suit your needs.

Prune
  • 76,765
  • 14
  • 60
  • 81