Maxtree example

This example shows how to use the component trees with Pylena

import matplotlib.pyplot as plt
from skimage.data import camera
import pylena as pln
import pandas as pd
import numpy as np

img = camera()

Compute the max-tree and the area of component and a custom attribute

maxtree = pln.morpho.maxtree(img, 4)
area = maxtree.compute_area()


mass_centers = maxtree.compute_attribute(lambda init, p: (init[0] + p[0], init[1] + p[1]), (0, 0))
mass_centers = np.array(list(mass_centers), ndmin=2)

df = pd.DataFrame(
    {
        "parent": maxtree.parent,
        "gray": maxtree.values,
        "area": area,
        "mc_x": mass_centers[:, 0] / area,
        "mc_y": mass_centers[:, 1] / area,
    }
)

df

	parent	gray	area	mc_x	mc_y
0	-1	0	262144	255.500000	255.500000
1	0	1	262143	255.499498	255.500525
2	1	2	262142	255.498993	255.501049
3	2	3	262122	255.492103	255.510003
4	3	4	261495	255.318052	255.834716
...	...	...	...	...	...
48994	3	4	1	328.000000	119.000000
48995	3	4	1	314.000000	152.000000
48996	3	4	1	308.000000	152.000000
48997	3	4	3	327.666667	136.666667
48998	3	4	1	329.000000	135.000000

48999 rows × 5 columns

Filtering and reconstruction

maxtree.filter((df["area"] > 200).values, inplace=True)
filtered = maxtree.reconstruct()

fig, axes = plt.subplots(nrows=1, ncols=2, figsize=(10, 5))
axes[0].imshow(img, cmap="gray")
axes[1].imshow(filtered, cmap="gray")
axes[0].set_title("Original")
axes[1].set_title("Filtered")
for i in range(2):
    axes[i].set_xticks([])
    axes[i].set_yticks([])
plt.tight_layout()

plt.show()