# Visualize
# First setup a 5-4-3 composite
def color_stretch(image, index, minmax=(0, 10000)):
colors = image[:, :, index].astype(np.float64)
max_val = minmax[1]
min_val = minmax[0]
# Enforce maximum and minimum values
colors[colors[:, :, :] > max_val] = max_val
colors[colors[:, :, :] < min_val] = min_val
for b in range(colors.shape[2]):
colors[:, :, b] = colors[:, :, b] * 1 / (max_val - min_val)
return colors
img543 = color_stretch(img, [4, 3, 2], (0, 8000))
# See https://github.com/matplotlib/matplotlib/issues/844/
n = class_prediction.max()
# Next setup a colormap for our map
colors = dict((
(0, (0, 0, 0, 255)), # Nodata
(1, (0, 150, 0, 255)), # Forest
(2, (0, 0, 255, 255)), # Water
(3, (0, 255, 0, 255)), # Herbaceous
(4, (160, 82, 45, 255)), # Barren
(5, (255, 0, 0, 255)) # Urban
))
# Put 0 - 255 as float 0 - 1
for k in colors:
v = colors[k]
_v = [_v / 255.0 for _v in v]
colors[k] = _v
index_colors = [colors[key] if key in colors else
(255, 255, 255, 0) for key in range(1, n + 1)]
cmap = plt.matplotlib.colors.ListedColormap(index_colors, 'Classification', n)
# Now show the classmap next to the image
plt.subplot(121)
plt.imshow(img543)
plt.subplot(122)
plt.imshow(class_prediction, cmap=cmap, interpolation='none')
plt.show()
RetroSearch is an open source project built by @garambo | Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4