forked from insarlab/MintPy
-
Notifications
You must be signed in to change notification settings - Fork 0
/
insar_vs_gps.py
419 lines (364 loc) · 18.4 KB
/
insar_vs_gps.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
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
############################################################
# Program is part of MintPy #
# Copyright (c) 2013, Zhang Yunjun, Heresh Fattahi #
# Author: Zhang Yunjun, 2018 #
############################################################
# Recommend import:
# from mintpy.objects.insar_vs_gps import insar_vs_gps
import sys
import numpy as np
from scipy import stats
from scipy.interpolate import griddata
from datetime import datetime as dt
from dateutil.relativedelta import relativedelta
from matplotlib import pyplot as plt
from mintpy.objects import timeseries, giantTimeseries
from mintpy.utils import ptime, readfile, plot as pp, utils as ut
from mintpy.objects.gps import GPS
from mintpy.defaults.plot import *
############################## utilities functions ##########################################
def plot_insar_vs_gps_scatter(vel_file, ref_gps_site, csv_file='gps_enu2los.csv', msk_file=None,
xname='InSAR', vlim=None, ex_gps_sites=[], display=True):
"""Scatter plot to compare the velocities between SAR/InSAR and GPS.
Parameters: vel_file - str, path of InSAR LOS velocity HDF5 file.
ref_gps_site - str, reference GNSS site name
csv_file - str, path of GNSS CSV file, generated after running view.py --gps-comp
msk_file - str, path of InSAR mask file.
xname - str, xaxis label
vlim - list of 2 float, display value range in the unit of cm/yr
Default is None to grab from data
If set, the range will be used to prune the SAR and GPS observations
ex_gps_sites - list of str, exclude GNSS sites for analysis and plotting.
Example:
from mintpy.objects.insar_vs_gps import plot_insar_vs_gps_scatter
csv_file = os.path.join(work_dir, 'geo/gps_enu2los.csv')
vel_file = os.path.join(work_dir, 'geo/geo_velocity.h5')
msk_file = os.path.join(work_dir, 'geo/geo_maskTempCoh.h5')
plot_insar_vs_gps_scatter(vel_file, ref_gps_site='CACT', csv_file=csv_file, msk_file=msk_file, vlim=[-2.5, 2])
"""
disp_unit = 'cm/yr'
unit_fac = 100.
# read GPS velocity from CSV file (generated by gps.get_gps_los_obs())
col_names = ['Site', 'Lon', 'Lat', 'Displacement', 'Velocity']
num_col = len(col_names)
col_types = ['U10'] + ['f8'] * (num_col - 1)
print('read GPS velocity from file: {}'.format(csv_file))
fc = np.genfromtxt(csv_file, dtype=col_types, delimiter=',', names=True)
sites = fc['Site']
lats = fc['Lat']
lons = fc['Lon']
gps_obs = fc[col_names[-1]] * unit_fac
if ex_gps_sites:
ex_flag = np.array([x in ex_gps_sites for x in sites], dtype=np.bool_)
if np.sum(ex_flag) > 0:
sites = sites[~ex_flag]
lats = lats[~ex_flag]
lons = lons[~ex_flag]
gps_obs = gps_obs[~ex_flag]
# read InSAR velocity
print('read InSAR velocity from file: {}'.format(vel_file))
atr = readfile.read_attribute(vel_file)
coord = ut.coordinate(atr)
ys, xs = coord.geo2radar(lats, lons)[:2]
if msk_file:
msk = readfile.read(msk_file)[0]
num_site = sites.size
insar_obs = np.zeros(num_site, dtype=np.float32) * np.nan
prog_bar = ptime.progressBar(maxValue=num_site)
for i in range(num_site):
x, y = xs[i], ys[i]
box = (x, y, x+1, y+1)
if msk_file and not msk[y, x]:
box = None
if box:
insar_obs[i] = readfile.read(vel_file, datasetName='velocity', box=box)[0] * unit_fac
prog_bar.update(i+1, suffix='{}/{} {}'.format(i+1, num_site, sites[i]))
prog_bar.close()
# reference site
ref_ind = sites.tolist().index(ref_gps_site)
gps_obs -= gps_obs[ref_ind]
insar_obs -= insar_obs[ref_ind]
# remove NaN value
print('removing sites with NaN values in GPS or {}'.format(xname))
flag = np.multiply(~np.isnan(insar_obs), ~np.isnan(gps_obs))
if vlim is not None:
print('pruning sites with value range: {} {}'.format(vlim, disp_unit))
flag *= gps_obs >= vlim[0]
flag *= gps_obs <= vlim[1]
flag *= insar_obs >= vlim[0]
flag *= insar_obs <= vlim[1]
gps_obs = gps_obs[flag]
insar_obs = insar_obs[flag]
sites = sites[flag]
# stats
print('GPS min/max: {:.2f} / {:.2f}'.format(np.nanmin(gps_obs), np.nanmax(gps_obs)))
print('InSAR min/max: {:.2f} / {:.2f}'.format(np.nanmin(insar_obs), np.nanmax(insar_obs)))
rmse = np.sqrt(np.sum((insar_obs - gps_obs)**2) / (gps_obs.size - 1))
r2 = stats.linregress(insar_obs, gps_obs)[2]
print('RMSE = {:.1f} cm'.format(rmse))
print('R^2 = {:.2f}'.format(r2))
# plot
if display:
plt.rcParams.update({'font.size': 12})
if vlim is None:
vlim = [np.min(insar_obs), np.max(insar_obs)]
buffer = (vlim[1] - vlim[0]) * 0.1
vlim = [vlim[0] - buffer, vlim[1] + buffer]
fig, ax = plt.subplots(figsize=[4, 4])
ax.plot((vlim[0], vlim[1]), (vlim[0], vlim[1]), 'k--')
ax.plot(insar_obs, gps_obs, '.', ms=15)
# axis format
ax.set_xlim(vlim)
ax.set_ylim(vlim)
ax.set_xlabel(f'{xname} [{disp_unit}]')
ax.set_ylabel(f'GNSS [{disp_unit}]')
ax.set_aspect('equal', 'box')
fig.tight_layout()
# output
out_fig = '{}_vs_gps_scatter.pdf'.format(xname.lower())
plt.savefig(out_fig, bbox_inches='tight', transparent=True, dpi=300)
print('save figure to file', out_fig)
plt.show()
return sites, insar_obs, gps_obs
############################## beginning of insar_vs_gps class ##############################
class insar_vs_gps:
""" Comparing InSAR time-series with GPS time-series in LOS direction
Parameters: ts_file : str, time-series HDF5 file
geom_file : str, geometry HDF5 file
temp_coh_file : str, temporal coherence HDF5 file
site_names : list of str, GPS site names
gps_dir : str, directory of the local GPS data files
ref_site : str, common reference site in space for InSAR and GPS
start/end_date : str, date in YYYYMMDD format for the start/end date
min_ref_date : str, date in YYYYMMDD format for the earliest common
reference date between InSAR and GPS
Returns: ds : dict, each element has the following components:
'GV03': {
'name': 'GV03',
'lat': -0.7977926892712729,
'lon': -91.13294444114553,
'gps_datetime': array([datetime.datetime(2014, 11, 1, 0, 0),
datetime.datetime(2014, 11, 2, 0, 0),
...,
datetime.datetime(2018, 6, 25, 0, 0)], dtype=object),
'gps_dis': array([-2.63673663e-02, ..., 6.43612206e-01], dtype=float32),
'gps_std': array([0.00496152, ..., 0.00477411], dtype=float32),
'reference_site': 'GV01',
'insar_datetime': array([datetime.datetime(2014, 12, 13, 0, 0),
datetime.datetime(2014, 12, 25, 0, 0),
...,
datetime.datetime(2018, 6, 19, 0, 0)], dtype=object),
'insar_dis_linear': array([-0.01476493, ..., 0.62273948]),
'temp_coh': 0.9961861392598478,
'gps_std_mean': 0.004515478,
'comm_dis_gps': array([-0.02635017, ..., 0.61315614], dtype=float32),
'comm_dis_insar': array([-0.01476493, ..., 0.60640174], dtype=float32),
'r_square': 0.9993494518609801,
'dis_rmse': 0.008023425326946351
}
"""
def __init__(self, ts_file, geom_file, temp_coh_file,
site_names, gps_dir='./GPS', ref_site='GV01',
start_date=None, end_date=None, min_ref_date=None):
self.insar_file = ts_file
self.geom_file = geom_file
self.temp_coh_file = temp_coh_file
self.site_names = site_names
self.gps_dir = gps_dir
self.ref_site = ref_site
self.num_site = len(site_names)
self.ds = {}
self.start_date = start_date
self.end_date = end_date
self.min_ref_date = min_ref_date
def open(self):
atr = readfile.read_attribute(self.insar_file)
k = atr['FILE_TYPE']
if k == 'timeseries':
ts_obj = timeseries(self.insar_file)
elif k == 'giantTimeseries':
ts_obj = giantTimeseries(self.insar_file)
else:
raise ValueError('Un-supported time-series file: {}'.format(k))
ts_obj.open(print_msg=False)
self.metadata = dict(ts_obj.metadata)
self.num_date = ts_obj.numDate
# remove time info from insar_datetime to be consistent with gps_datetime
self.insar_datetime = np.array([i.replace(hour=0, minute=0, second=0, microsecond=0)
for i in ts_obj.times])
# default start/end
if self.start_date is None:
self.start_date = (ts_obj.times[0] - relativedelta(months=1)).strftime('%Y%m%d')
if self.end_date is None:
self.end_date = (ts_obj.times[-1] + relativedelta(months=1)).strftime('%Y%m%d')
# default min_ref_date
if self.min_ref_date is None:
self.min_ref_date = ts_obj.times[5].strftime('%Y%m%d')
elif self.min_ref_date not in ts_obj.dateList:
raise ValueError('input min_ref_date {} does not exist in insar file {}'.format(
self.min_ref_date, self.insar_file))
self.read_gps()
self.read_insar()
self.calculate_rmse()
return
def read_gps(self):
for sname in self.site_names:
site = {}
site['name'] = sname
gps_obj = GPS(sname, data_dir=self.gps_dir)
gps_obj.open(print_msg=False)
site['lat'] = gps_obj.site_lat
site['lon'] = gps_obj.site_lon
(site['gps_datetime'],
site['gps_dis'],
site['gps_std']) = gps_obj.read_gps_los_displacement(self.geom_file, self.start_date, self.end_date,
ref_site=self.ref_site,
gps_comp='enu2los')[0:3]
site['reference_site'] = self.ref_site
self.ds[sname] = site
sys.stdout.write('\rreading GPS {}'.format(sname))
sys.stdout.flush()
print()
return
def read_insar(self):
# 2.1 prepare interpolation
coord = ut.coordinate(self.metadata, lookup_file=self.geom_file)
lats = [self.ds[k]['lat'] for k in self.ds.keys()]
lons = [self.ds[k]['lon'] for k in self.ds.keys()]
geo_box = (min(lons), max(lats), max(lons), min(lats)) #(W, N, E, S)
pix_box = coord.bbox_geo2radar(geo_box) #(400, 1450, 550, 1600)
src_lat = readfile.read(self.geom_file, datasetName='latitude', box=pix_box)[0].reshape(-1,1)
src_lon = readfile.read(self.geom_file, datasetName='longitude', box=pix_box)[0].reshape(-1,1)
src_pts = np.hstack((src_lat, src_lon))
dest_pts = np.zeros((self.num_site, 2))
for i in range(self.num_site):
site = self.ds[self.site_names[i]]
dest_pts[i,:] = site['lat'], site['lon']
# 2.2 interpolation - displacement / temporal coherence
interp_method = 'linear' #nearest, linear, cubic
src_value, atr = readfile.read(self.insar_file, box=pix_box)
src_value = src_value.reshape(self.num_date, -1)
if atr['FILE_TYPE'] == 'giantTimeseries':
src_value *= 0.001
insar_dis = np.zeros((self.num_site, self.num_date))
for i in range(self.num_date):
insar_dis[:,i] = griddata(src_pts, src_value[i,:], dest_pts, method=interp_method)
sys.stdout.write(('\rreading InSAR acquisition {}/{}'
' with {} interpolation').format(i+1, self.num_date, interp_method))
sys.stdout.flush()
print()
print('reading temporal coherence')
src_value = readfile.read(self.temp_coh_file, box=pix_box)[0].flatten()
temp_coh = griddata(src_pts, src_value, dest_pts, method=interp_method)
# 2.3 write interpolation result
self.insar_dis_name = 'insar_dis_{}'.format(interp_method)
insar_dis_ref = insar_dis[self.site_names.index(self.ref_site),:]
for i in range(self.num_site):
site = self.ds[self.site_names[i]]
site['insar_datetime'] = self.insar_datetime
site[self.insar_dis_name] = insar_dis[i,:] - insar_dis_ref # reference insar to the precise location in space
site['temp_coh'] = temp_coh[i]
# 2.4 reference insar and gps to a common date
print('reference insar and gps to a common date')
for i in range(self.num_site):
site = self.ds[self.site_names[i]]
gps_date = site['gps_datetime']
insar_date = site['insar_datetime']
# find common reference date
ref_date = dt.strptime(self.min_ref_date, "%Y%m%d")
ref_idx = insar_date.tolist().index(ref_date)
while ref_idx < self.num_date:
if insar_date[ref_idx] not in gps_date:
ref_idx += 1
else:
break
if ref_idx == self.num_date:
raise RuntimeError('InSAR and GPS do not share ANY date for site: {}'.format(site['name']))
comm_date = insar_date[ref_idx]
# reference insar in time
site[self.insar_dis_name] -= site[self.insar_dis_name][ref_idx]
# reference gps dis/std in time
ref_idx_gps = np.where(gps_date == comm_date)[0][0]
site['gps_dis'] -= site['gps_dis'][ref_idx_gps]
site['gps_std'] = np.sqrt(site['gps_std']**2 + site['gps_std'][ref_idx_gps]**2)
site['gps_std_mean'] = np.mean(site['gps_std'])
return
def calculate_rmse(self):
## 3. calculate RMSE
for i in range(self.num_site):
site = self.ds[self.site_names[i]]
gps_date = site['gps_datetime']
insar_date = site['insar_datetime']
comm_dates = np.array(sorted(list(set(gps_date) & set(insar_date))))
num_comm_date = len(comm_dates)
# get displacement at common dates
comm_dis_insar = np.zeros(num_comm_date, np.float32)
comm_dis_gps = np.zeros(num_comm_date, np.float32)
for j in range(num_comm_date):
idx1 = np.where(gps_date == comm_dates[j])[0][0]
idx2 = np.where(insar_date == comm_dates[j])[0][0]
comm_dis_gps[j] = site['gps_dis'][idx1]
comm_dis_insar[j] = site[self.insar_dis_name][idx2]
site['comm_dis_gps'] = comm_dis_gps
site['comm_dis_insar'] = comm_dis_insar
site['r_square'] = stats.linregress(comm_dis_gps, comm_dis_insar)[2]
site['dis_rmse'] = np.sqrt(np.sum(np.square(comm_dis_gps - comm_dis_insar)) / (num_comm_date - 1))
#print('site: {}, RMSE: {:.1f} cm'.format(self.site_names[i], dis_rmse*100.))
def sort_by_velocity(ds):
## 4. calculate velocity to sort plotting order
site_vel = {}
site_names = sorted(list(ds.keys()))
for sname in site_names:
site = ds[sname]
# design matrix
yr_diff = np.array([i.year + (i.timetuple().tm_yday - 1) / 365.25 for i in site['gps_datetime']])
yr_diff -= yr_diff[0]
A = np.ones([len(site['gps_datetime']), 2], dtype=np.float32)
A[:, 0] = yr_diff
# LS estimation
ts = np.array(site['gps_dis'])
ts -= ts[0]
X = np.dot(np.linalg.pinv(A), ts)[0]
site_vel[sname] = X
site_names2plot = [i[0] for i in sorted(site_vel.items(), key=lambda kv: kv[1], reverse=True)]
site_names2plot = [i for i in site_names2plot if site_vel[i] != 0]
return site_names2plot
def print_stats(ds):
site_names = sorted(list(ds.keys()))
for sname in site_names:
site = ds[sname]
print('{}, rmse: {:.1f} cm, r_square: {:.2f}, temp_coh: {:.2f}'.format(sname,
site['dis_rmse']*100.,
site['r_square'],
site['temp_coh']))
return
def plot_one_site(ax, site, offset=0.):
# GPS
ax.errorbar(site['gps_datetime'],
site['gps_dis']-offset,
yerr=site['gps_std']*3.,
ms=marker_size*0.2, lw=0, alpha=1., fmt='-o',
elinewidth=edge_width*0.5, ecolor=pp.mplColors[0],
capsize=marker_size*0.25, markeredgewidth=edge_width*0.5,
label='GPS', zorder=1)
# InSAR
if site['temp_coh'] < 0.7:
ecolor = 'gray'
else:
ecolor = pp.mplColors[1]
insar_dis_name = [i for i in site.keys() if i.startswith('insar_dis')][0]
ax.scatter(site['insar_datetime'],
site[insar_dis_name]-offset,
s=5**2, label='InSAR',
facecolors='none', edgecolors=ecolor, linewidth=1., alpha=0.7, zorder=2)
# Label
ax.annotate('{:.1f} / {:.2f} / {:.2f}'.format(site['dis_rmse']*100., site['r_square'], site['temp_coh']),
xy=(1.03, site[insar_dis_name][-1] - offset - 0.02),
xycoords=ax.get_yaxis_transform(), # y in data untis, x in axes fraction
color='k', fontsize=font_size)
ax.annotate('{}'.format(site['name']),
xy=(0.05, site[insar_dis_name][0] - offset + 0.1),
xycoords=ax.get_yaxis_transform(), # y in data untis, x in axes fraction
color='k', fontsize=font_size)
return ax
############################## end of insar_vs_gps class ####################################