# prune

MainPage

Description: This method implements the pruning strategy.

Signature: MorphTreeAlpha.prune(to_prune)

Input:
• to_prune, 1d-array bool. Array indicating whcih nodes should be pruned.

Output:

# C++ Aux Function

```OK
```
```OK [C/C++ extension is up-to-date]
```

# Python Implementation

``` 1 import numpy as np
2
3 def prune(self, to_prune):
4     """
5     Contracts entire branches of the tree. This is the prunning procedure to be
6     used with increasing connected filters. If a node is indicated in to_prune, all its descendants
7     should also be indicated in to_prune.
8     """
9     N = self.node_array.shape[1]
10     lut = np.arange(N, dtype = np.int32)
11     to_prune[0] = False
12     self.prune_aux(lut,to_prune.astype(np.int32), self.node_array[0,:], self.node_array[1,:])
13     #self.node_index = lut[self.node_index]
14     self.compact(to_prune,lut)
15     return self
```

# Example

``` 1 from morph_tree_alpha import MorphTreeAlpha
2 import numpy as np
3
4 img = np.array([[100, 100, 100, 0,   0,   0,   0,   0],\
5                 [150, 150, 150, 150, 150, 150, 150, 150],\
6                 [150, 160, 190, 150, 200, 200, 229, 150], \
7                 [150, 185, 255, 150, 200, 230, 200, 150],\
8                 [150, 180, 200, 150, 215, 229, 200, 150],\
9                 [150, 150, 150, 150, 150, 150, 150, 150],\
10                 [50,  50,  0,   0,   0,   0,   0,   0]], dtype = np.uint8)
11
12
13 Bc = np.array([[0,1,0],\
14                [1,1,1],\
15                [0,1,0]], dtype = bool)
16
17
18 morph_tree = MorphTreeAlpha(img,Bc)
19
20
21 to_prune = morph_tree.node_array[3,:] < 6
22
23
24 g= morph_tree.generateGraph(keep = ~to_prune)
25 mmgraphviz(g, title='Morph-tree and its nodes marked to be pruned')
26
27 morph_tree.prune(to_prune)
28
29
30
31 g= morph_tree.generateGraph()
32 mmgraphviz(g, title='Morph-tree after the filtering procedure')
```