Here is my question
* value is a 2-d np.array(31 x 37) with value 0,1.
It shows like this:
Only plot the grid in blue color if the value == 1
value_mask = np.ma.masked_less(value[:,:],0.001)
pc =plt.pcolor(xx,yy,value_mask,alpha =1,facecolor = "pink",edgecolor = 'steelblue',zorder =3)
How to sum the amount of the contiguous "blue grid".
In this case, I mean the 1 upper right grid isolated from the majority will be ignored.
You can use scipy.ndimage.label to label contiguous regions in your value array, then sum the number of elements for each label:
import numpy as np
from scipy import ndimage
value = np.array([[0, 0, 1, 1, 0, 0],
[0, 0, 0, 1, 0, 0],
[1, 1, 0, 0, 1, 0],
[0, 0, 0, 1, 0, 0]])
# label contiguous regions of non-zero elements
labels, nfeatures = ndimage.label(value)
# sizes of each region
sizes = (labels == (np.arange(nfeatures) + 1)[:, None, None]).sum((1, 2))
biggest = sizes.max() # number of non-zero elements in largest contiguous region
print(biggest)
# 3