/
dc_fractional_coverage_classifier.py
266 lines (219 loc) · 9.86 KB
/
dc_fractional_coverage_classifier.py
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
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
import numpy as np
import xarray as xr
import scipy.optimize as opt #nnls
import datacube
from . import dc_utilities as utilities
from .dc_utilities import create_default_clean_mask
from .clean_mask import create_2D_mosaic_clean_mask
# Command line tool imports
import argparse
import os
import collections
import gdal
from datetime import datetime
# Author: KMF
# Creation date: 2016-10-24
def frac_coverage_classify(dataset_in, clean_mask=None, no_data=-9999):
"""
Description:
Performs fractional coverage algorithm on given dataset.
Assumption:
- The implemented algorithm is defined for Landsat 5/Landsat 7; in order for it to
be used for Landsat 8, the bands will need to be adjusted
References:
- Guerschman, Juan P., et al. "Assessing the effects of site heterogeneity and soil
properties when unmixing photosynthetic vegetation, non-photosynthetic vegetation
and bare soil fractions from Landsat and MODIS data." Remote Sensing of Environment
161 (2015): 12-26.
-----
Inputs:
dataset_in (xarray.Dataset) - dataset retrieved from the Data Cube - should be a mosaic (mean mosaic is a good choice) containing:
coordinates: latitude, longitude
variables: blue, green, red, nir, swir1, swir2
Optional Inputs:
clean_mask (numpy.ndarray with dtype boolean) - true for values user considers clean;
if user does not provide a clean mask, all values will be considered clean
no_data (int/float) - no data pixel value; default: -9999
Output:
dataset_out (xarray.Dataset) - fractional coverage results with no data = -9999; containing
coordinates: latitude, longitude
variables: bs, pv, npv
where bs -> bare soil, pv -> photosynthetic vegetation, npv -> non-photosynthetic vegetation
"""
# Default to masking nothing.
if clean_mask is None:
clean_mask = create_default_clean_mask(dataset_in)
# Clean and format data so it is analysis-ready.
band_stack = []
mosaic_clean_mask = create_2D_mosaic_clean_mask(clean_mask)
mosaic_clean_mask_flat = mosaic_clean_mask.flatten()
for band in [
dataset_in.blue.values, dataset_in.green.values, dataset_in.red.values, dataset_in.nir.values,
dataset_in.swir1.values, dataset_in.swir2.values
]:
band = band.astype(np.float32)
band = band * 0.0001
band = band.flatten()
band_clean = np.full(band.shape, np.nan)
band_clean[mosaic_clean_mask_flat] = band[mosaic_clean_mask_flat]
band_stack.append(band_clean)
band_stack = np.array(band_stack).transpose()
num_bands = band_stack.shape[1]
# In order to account for the non-linearities in the spectral mixing, the following
# code performs log transformations of the surface reflectance bands and interactive
# terms in the regression equations.
for b in range(num_bands):
band_stack = np.hstack((band_stack, np.expand_dims(np.log(band_stack[:, b]), axis=1)))
for b in range(num_bands):
band_stack = np.hstack(
(band_stack, np.expand_dims(np.multiply(band_stack[:, b], band_stack[:, b + num_bands]), axis=1)))
for b in range(num_bands):
for b2 in range(b + 1, num_bands):
band_stack = np.hstack(
(band_stack, np.expand_dims(np.multiply(band_stack[:, b], band_stack[:, b2]), axis=1)))
for b in range(num_bands):
for b2 in range(b + 1, num_bands):
band_stack = np.hstack(
(band_stack, np.expand_dims(np.multiply(band_stack[:, b + num_bands], band_stack[:, b2 + num_bands]), axis=1)))
for b in range(num_bands):
for b2 in range(b + 1, num_bands):
band_stack = np.hstack((band_stack, np.expand_dims(
np.divide(band_stack[:, b2] - band_stack[:, b], band_stack[:, b2] + band_stack[:, b]), axis=1)))
band_stack = np.nan_to_num(band_stack) # Now a n x 63 matrix (assuming one acquisition)
# Run fractional coverage algorithm.
ones = np.ones(band_stack.shape[0])
ones = ones.reshape(ones.shape[0], 1)
band_stack = np.concatenate((band_stack, ones), axis=1) # Now a n x 64 matrix (assuming one acquisition)
end_members = np.loadtxt(
'/home/localuser/Datacube/data_cube_ui/utils/data_cube_utilities/endmembers_landsat.csv',
delimiter=',') # Creates a 64 x 3 matrix
SumToOneWeight = 0.02
ones = np.ones(end_members.shape[1]) * SumToOneWeight
ones = ones.reshape(1, end_members.shape[1])
end_members = np.concatenate((end_members, ones), axis=0).astype(np.float32)
result = np.zeros((band_stack.shape[0], end_members.shape[1]), dtype=np.float32) # Creates an n x 3 matrix
for i in range(band_stack.shape[0]):
if mosaic_clean_mask_flat[i]:
result[i, :] = (opt.nnls(end_members, band_stack[i, :])[0].clip(0, 2.54) * 100).astype(np.int16)
else:
result[i, :] = np.ones((end_members.shape[1]), dtype=np.int16) * (-9999) # Set as no data
latitude = dataset_in.latitude
longitude = dataset_in.longitude
result = result.reshape(latitude.size, longitude.size, 3)
# Photosynthetic Vegetation
pv_band = result[:, :, 0]
# Nonphotosynthetic Vegetation
npv_band = result[:, :, 1]
# Bare Soil
bs_band = result[:, :, 2]
pv_clean = np.full(pv_band.shape, -9999)
npv_clean = np.full(npv_band.shape, -9999)
bs_clean = np.full(bs_band.shape, -9999)
pv_clean[mosaic_clean_mask] = pv_band[mosaic_clean_mask]
npv_clean[mosaic_clean_mask] = npv_band[mosaic_clean_mask]
bs_clean[mosaic_clean_mask] = bs_band[mosaic_clean_mask]
rapp_bands = collections.OrderedDict([('bs', (['latitude', 'longitude'], bs_band)),
('pv', (['latitude', 'longitude'], pv_band)),
('npv', (['latitude', 'longitude'], npv_band))])
rapp_dataset = xr.Dataset(rapp_bands, coords={'latitude': latitude, 'longitude': longitude})
return rapp_dataset
def main(platform, product_type, min_lon, max_lon, min_lat, max_lat, start_date, end_date, dc_config):
"""
Description:
Command-line fractional coverage tool - TODO
Assumptions:
The command-line tool assumes there is a measurement called cf_mask
Inputs:
platform (str)
product_type (str)
min_lon (str)
max_lon (str)
min_lat (str)
max_lat (str)
start_date (str)
end_date (str)
dc_config (str)
"""
# Initialize data cube object
dc = datacube.Datacube(config=dc_config, app='dc-frac-cov')
products = dc.list_products()
platform_names = set([product[6] for product in products.values])
if platform not in platform_names:
print('ERROR: Invalid platform.')
print('Valid platforms are:')
for name in platform_names:
print(name)
return
product_names = [product[0] for product in products.values]
if product_type not in product_names:
print('ERROR: Invalid product type.')
print('Valid product types are:')
for name in product_names:
print(name)
return
try:
min_lon = float(args.min_lon)
max_lon = float(args.max_lon)
min_lat = float(args.min_lat)
max_lat = float(args.max_lat)
except:
print('ERROR: Longitudes/Latitudes must be float values')
return
try:
start_date_str = start_date
end_date_str = end_date
start_date = datetime.strptime(start_date, '%Y-%m-%d')
end_date = datetime.strptime(end_date, '%Y-%m-%d')
except:
print('ERROR: Invalid date format. Date format: YYYY-MM-DD')
return
if not os.path.exists(dc_config):
print('ERROR: Invalid file path for dc_config')
return
# Retrieve data from Data Cube
dataset_in = dc.load(
platform=platform,
product=product_type,
time=(start_date, end_date),
lon=(min_lon, max_lon),
lat=(min_lat, max_lat))
# Get information needed for saving as GeoTIFF
# Spatial ref
crs = dataset_in.crs
spatial_ref = utilities.get_spatial_ref(crs)
# Upper left coordinates
ul_lon = dataset_in.longitude.values[0]
ul_lat = dataset_in.latitude.values[0]
# Resolution
products = dc.list_products()
resolution = products.resolution[products.name == 'ls7_ledaps']
lon_dist = resolution.values[0][1]
lat_dist = resolution.values[0][0]
# Rotation
lon_rtn = 0
lat_rtn = 0
geotransform = (ul_lon, lon_dist, lon_rtn, ul_lat, lat_rtn, lat_dist)
dataset_out = frac_coverage_classify(dataset_in)
out_file = (str(min_lon) + '_' + str(min_lat) + '_' + start_date_str + '_' + end_date_str + '_frac_coverage.tif')
utilities.save_to_geotiff(out_file, gdal.GDT_Float32, dataset_out, geotransform, spatial_ref)
if __name__ == '__main__':
start_time = datetime.now()
parser = argparse.ArgumentParser()
parser.add_argument('platform', help='Data platform; example: LANDSAT_7')
parser.add_argument('product', help='Product type; example: ls7_ledaps')
parser.add_argument('min_lon', help='Minimum longitude')
parser.add_argument('max_lon', help='Maximum longitude')
parser.add_argument('min_lat', help='Minimum latitude')
parser.add_argument('max_lat', help='Maximum latitude')
parser.add_argument('start_date', help='Start date; format: YYYY-MM-DD')
parser.add_argument('end_date', help='End date; format: YYYY-MM-DD')
parser.add_argument(
'dc_config',
nargs='?',
default='~/.datacube.conf',
help='Datacube configuration path; default: ~/.datacube.conf')
args = parser.parse_args()
main(args.platform, args.product, args.min_lon, args.max_lon, args.min_lat, args.max_lat, args.start_date,
args.end_date, args.dc_config)
end_time = datetime.now()
print('Execution time: ' + str(end_time - start_time))