classification/Cloudmasking.py at master · MapTailor/classification · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
from sklearn.cluster import MiniBatchKMeans

import skimage.io as io
import numpy as np
import os, sys
import json
from skimage.morphology import disk
from skimage.morphology import dilation

def cloudmasking(self,):

    band = os.path.join(self.workdir, band + ".tif")

    img_ds = io.imread(band)
    img = np.array(img_ds, dtype='uint16')

    row = img.shape[0]
    col = img.shape[1]

    ras_shape = (row * col, 1)
    img_as_array = img[:, :].reshape(ras_shape)

    cluster = MiniBatchKMeans(n_clusters=2,
                              tol=0.00001,
                              n_init=10,
                              init='random',
                              max_no_improvement=100,
                              reassignment_ratio=0.05)

    cluster_labels = cluster.fit_predict(img_as_array)
    cluster_center = cluster.cluster_centers_

    new_shape = (row, col)
    img_clusters = cluster_labels.reshape(new_shape)

    centroids_num = len(cluster_center)

    liste = []
    for i in range(0, centroids_num):

        means = np.mean(cluster_center[i])
        liste.append(means)

    max = np.max(liste)
    val = liste.index(max)

    if max > 1800:

        mask = np.where(img_clusters == val, 1, np.nan)
        img_mask = np.array(mask, dtype='uint8').reshape((row, col))

        kernel = disk(5)
        img_buffer = dilation(img_mask, kernel)

        cloudmask = os.path.join(self.workdir, "cloudmask.tif").format(sys.platform, os.sep)
        io.imsave(cloudmask, img_buffer)