|
1
|
|
|
import sys |
|
2
|
|
|
import os |
|
3
|
|
|
import math |
|
4
|
|
|
|
|
5
|
|
|
from timeit import default_timer as timer |
|
6
|
|
|
from multiprocessing import Process, Queue |
|
7
|
|
|
|
|
8
|
|
|
import numpy |
|
9
|
|
|
from scipy.spatial import KDTree |
|
10
|
|
|
|
|
11
|
|
|
import matplotlib.pyplot as plt |
|
12
|
|
|
from matplotlib.figure import Figure |
|
13
|
|
|
from matplotlib.backends.backend_agg import FigureCanvasAgg as FigureCanvas |
|
14
|
|
|
|
|
15
|
|
|
import geojson |
|
16
|
|
|
import geojsoncontour |
|
17
|
|
|
import togeojsontiles |
|
18
|
|
|
|
|
19
|
|
|
import nsmaps |
|
20
|
|
|
from nsmaps.logger import logger |
|
21
|
|
|
|
|
22
|
|
|
|
|
23
|
|
|
TIPPECANOE_DIR = '/usr/local/bin/' |
|
24
|
|
|
|
|
25
|
|
|
|
|
26
|
|
|
def dotproduct(v1, v2): |
|
27
|
|
|
return sum((a * b) for a, b in zip(v1, v2)) |
|
28
|
|
|
|
|
29
|
|
|
def length(v): |
|
30
|
|
|
return math.sqrt(dotproduct(v, v)) |
|
31
|
|
|
|
|
32
|
|
|
def angle(v1, v2): |
|
33
|
|
|
return math.acos(dotproduct(v1, v2) / (length(v1) * length(v2))) |
|
34
|
|
|
|
|
35
|
|
|
|
|
36
|
|
|
class ContourData(object): |
|
37
|
|
|
def __init__(self): |
|
38
|
|
|
self.Z = None |
|
39
|
|
|
self.index_begin = 0 |
|
40
|
|
|
|
|
41
|
|
|
|
|
42
|
|
|
class ContourPlotConfig(object): |
|
43
|
|
|
def __init__(self): |
|
44
|
|
|
self.stepsize_deg = 0.005 |
|
45
|
|
|
self.n_processes = 4 |
|
46
|
|
|
self.cycle_speed_kmh = 18.0 |
|
47
|
|
|
self.n_nearest = 20 |
|
48
|
|
|
self.lon_start = 3.0 |
|
49
|
|
|
self.lat_start = 50.5 |
|
50
|
|
|
self.delta_deg = 6.5 |
|
51
|
|
|
self.lon_end = self.lon_start + self.delta_deg |
|
52
|
|
|
self.lat_end = self.lat_start + self.delta_deg / 2.0 |
|
53
|
|
|
self.min_angle_between_segments = 7 |
|
54
|
|
|
|
|
55
|
|
|
def print_bounding_box(self): |
|
56
|
|
|
print( |
|
57
|
|
|
'[[' + str(self.lon_start) + ',' + str(self.lat_start) + '],' \ |
|
58
|
|
|
'[' + str(self.lon_start) + ',' + str(self.lat_end) + '],' \ |
|
59
|
|
|
'[' + str(self.lon_end) + ',' + str(self.lat_end) + '],' \ |
|
60
|
|
|
'[' + str(self.lon_end) + ',' + str(self.lat_start) + '],' \ |
|
61
|
|
|
'[' + str(self.lon_start) + ',' + str(self.lat_start) + ']]' \ |
|
62
|
|
|
) |
|
63
|
|
|
|
|
64
|
|
|
|
|
65
|
|
|
class TestConfig(ContourPlotConfig): |
|
66
|
|
|
def __init__(self): |
|
67
|
|
|
super().__init__() |
|
68
|
|
|
self.stepsize_deg = 0.005 |
|
69
|
|
|
self.n_processes = 4 |
|
70
|
|
|
self.lon_start = 4.8 |
|
71
|
|
|
self.lat_start = 52.0 |
|
72
|
|
|
self.delta_deg = 1.0 |
|
73
|
|
|
self.lon_end = self.lon_start + self.delta_deg |
|
74
|
|
|
self.lat_end = self.lat_start + self.delta_deg / 2.0 |
|
75
|
|
|
self.min_angle_between_segments = 7 |
|
76
|
|
|
self.latrange = [] |
|
77
|
|
|
self.lonrange = [] |
|
78
|
|
|
self.Z = [[]] |
|
79
|
|
|
|
|
80
|
|
|
|
|
81
|
|
|
class Contour(object): |
|
82
|
|
|
def __init__(self, departure_station, stations, config, data_dir): |
|
83
|
|
|
self.departure_station = departure_station |
|
84
|
|
|
self.stations = stations |
|
85
|
|
|
self.config = config |
|
86
|
|
|
self.data_dir = data_dir |
|
87
|
|
|
|
|
88
|
|
|
def create_contour_data(self, filepath): |
|
89
|
|
|
if self.departure_station.has_travel_time_data(): |
|
90
|
|
|
self.stations.travel_times_from_json(self.departure_station.get_travel_time_filepath()) |
|
91
|
|
|
if os.path.exists(filepath): |
|
92
|
|
|
logger.error('Output file ' + filepath + ' already exists. Will not override.') |
|
93
|
|
|
return |
|
94
|
|
|
else: |
|
95
|
|
|
logger.error('Input file ' + self.departure_station.get_travel_time_filepath() + ' not found. Skipping station.') |
|
96
|
|
|
|
|
97
|
|
|
start = timer() |
|
98
|
|
|
numpy.set_printoptions(3, threshold=100, suppress=True) # .3f |
|
99
|
|
|
|
|
100
|
|
|
altitude = 0.0 |
|
101
|
|
|
self.lonrange = numpy.arange(self.config.lon_start, self.config.lon_end, self.config.stepsize_deg) |
|
102
|
|
|
self.latrange = numpy.arange(self.config.lat_start, self.config.lat_end, self.config.stepsize_deg / 2.0) |
|
103
|
|
|
self.Z = numpy.zeros((int(self.lonrange.shape[0]), int(self.latrange.shape[0]))) |
|
104
|
|
|
gps = nsmaps.utilgeo.GPS() |
|
105
|
|
|
|
|
106
|
|
|
positions = [] |
|
107
|
|
|
for station in self.stations: |
|
108
|
|
|
x, y, z = gps.lla2ecef([station.get_lat(), station.get_lon(), altitude]) |
|
109
|
|
|
positions.append([x, y, z]) |
|
110
|
|
|
|
|
111
|
|
|
logger.info('starting spatial interpolation') |
|
112
|
|
|
|
|
113
|
|
|
# tree to find nearest neighbors |
|
114
|
|
|
tree = KDTree(positions) |
|
115
|
|
|
|
|
116
|
|
|
queue = Queue() |
|
117
|
|
|
processes = [] |
|
118
|
|
|
if self.config.n_nearest > len(self.stations): |
|
119
|
|
|
self.config.n_nearest = len(self.stations) |
|
120
|
|
|
for i in range(0, self.config.n_processes): |
|
121
|
|
|
begin = i * len(self.latrange)/self.config.n_processes |
|
122
|
|
|
end = (i+1)*len(self.latrange)/self.config.n_processes |
|
123
|
|
|
latrange_part = self.latrange[begin:end] |
|
124
|
|
|
process = Process(target=self.interpolate_travel_time, args=(queue, i, self.stations.stations, tree, gps, latrange_part, |
|
125
|
|
|
self.lonrange, altitude, self.config.n_nearest, self.config.cycle_speed_kmh)) |
|
126
|
|
|
processes.append(process) |
|
127
|
|
|
|
|
128
|
|
|
for process in processes: |
|
129
|
|
|
process.start() |
|
130
|
|
|
|
|
131
|
|
|
# get from the queue and append the values |
|
132
|
|
|
for i in range(0, self.config.n_processes): |
|
133
|
|
|
data = queue.get() |
|
134
|
|
|
index_begin = data.index_begin |
|
135
|
|
|
begin = int(index_begin*len(self.latrange)/self.config.n_processes) |
|
136
|
|
|
end = int((index_begin+1)*len(self.latrange)/self.config.n_processes) |
|
137
|
|
|
self.Z[0:][begin:end] = data.Z |
|
138
|
|
|
|
|
139
|
|
|
for process in processes: |
|
140
|
|
|
process.join() |
|
141
|
|
|
|
|
142
|
|
|
end = timer() |
|
143
|
|
|
logger.info('finished spatial interpolation in ' + str(end - start) + ' [sec]') |
|
144
|
|
|
|
|
145
|
|
|
# self.create_geojson(filepath, max_zoom, min_zoom, stroke_width) |
|
146
|
|
|
|
|
147
|
|
|
def create_geojson(self, filepath, min_zoom=0, max_zoom=12, stroke_width=1, n_contours=41): |
|
148
|
|
|
figure = Figure(frameon=False) |
|
149
|
|
|
FigureCanvas(figure) |
|
150
|
|
|
ax = figure.add_subplot(111) |
|
151
|
|
|
levels = numpy.linspace(0, 200, num=n_contours) |
|
152
|
|
|
# contours = plt.contourf(lonrange, latrange, Z, levels=levels, cmap=plt.cm.plasma) |
|
153
|
|
|
contours = ax.contour(self.lonrange, self.latrange, self.Z, levels=levels, cmap=plt.cm.jet, linewidths=3) |
|
154
|
|
|
# cbar = figure.colorbar(contours, format='%.1f') |
|
155
|
|
|
# plt.savefig('contour_example.png', dpi=150) |
|
156
|
|
|
ndigits = len(str(int(1.0 / self.config.stepsize_deg))) + 1 |
|
157
|
|
|
logger.info('converting contour to geojson file: ' + filepath) |
|
158
|
|
|
geojsoncontour.contour_to_geojson( |
|
159
|
|
|
contour=contours, |
|
160
|
|
|
geojson_filepath=filepath, |
|
161
|
|
|
contour_levels=levels, |
|
162
|
|
|
min_angle_deg=self.config.min_angle_between_segments, |
|
163
|
|
|
ndigits=ndigits, |
|
164
|
|
|
unit='min', |
|
165
|
|
|
stroke_width=stroke_width |
|
166
|
|
|
) |
|
167
|
|
|
with open(filepath, 'r') as jsonfile: |
|
168
|
|
|
feature_collection = geojson.load(jsonfile) |
|
169
|
|
|
for feature in feature_collection['features']: |
|
170
|
|
|
feature["tippecanoe"] = {"maxzoom": str(int(max_zoom)), "minzoom": str(int(min_zoom))} |
|
171
|
|
|
dump = geojson.dumps(feature_collection, sort_keys=True) |
|
172
|
|
|
with open(filepath, 'w') as fileout: |
|
173
|
|
|
fileout.write(dump) |
|
174
|
|
|
|
|
175
|
|
|
cbar = figure.colorbar(contours, format='%d', orientation='horizontal') |
|
176
|
|
|
cbar.set_label('Travel time [minutes]') |
|
177
|
|
|
# cbar.set_ticks(self.config.colorbar_ticks) |
|
178
|
|
|
ax.set_visible(False) |
|
179
|
|
|
figure.savefig( |
|
180
|
|
|
filepath.replace('.geojson', '') + "_colorbar.png", |
|
181
|
|
|
dpi=90, |
|
182
|
|
|
bbox_inches='tight', |
|
183
|
|
|
pad_inches=0, |
|
184
|
|
|
transparent=True, |
|
185
|
|
|
) |
|
186
|
|
|
|
|
187
|
|
|
def create_geojson_tiles(self, filepaths, tile_dir, min_zoom=0, max_zoom=12): |
|
188
|
|
|
bound_box_filepath = os.path.join(self.data_dir, 'bounding_box.geojson') |
|
189
|
|
|
assert os.path.exists(bound_box_filepath) |
|
190
|
|
|
filepaths.append(bound_box_filepath) |
|
191
|
|
|
togeojsontiles.geojson_to_mbtiles( |
|
192
|
|
|
filepaths=filepaths, |
|
193
|
|
|
tippecanoe_dir=TIPPECANOE_DIR, |
|
194
|
|
|
mbtiles_file='out.mbtiles', |
|
195
|
|
|
minzoom=min_zoom, |
|
196
|
|
|
maxzoom=max_zoom, |
|
197
|
|
|
full_detail=10, |
|
198
|
|
|
lower_detail=9, |
|
199
|
|
|
min_detail=7 |
|
200
|
|
|
) |
|
201
|
|
|
logger.info('converting mbtiles to geojson-tiles') |
|
202
|
|
|
togeojsontiles.mbtiles_to_geojsontiles( |
|
203
|
|
|
tippecanoe_dir=TIPPECANOE_DIR, |
|
204
|
|
|
tile_dir=tile_dir, |
|
205
|
|
|
mbtiles_file='out.mbtiles', |
|
206
|
|
|
) |
|
207
|
|
|
logger.info('DONE: create contour json tiles') |
|
208
|
|
|
|
|
209
|
|
|
@staticmethod |
|
210
|
|
|
def interpolate_travel_time(q, position, stations, kdtree, gps, latrange, lonrange, altitude, n_nearest, cycle_speed_kmh): |
|
211
|
|
|
# n_nearest: check N nearest stations as best start for cycle route |
|
212
|
|
|
logger.info('interpolate_travel_time') |
|
213
|
|
|
Z = numpy.zeros((int(latrange.shape[0]), int(lonrange.shape[0]))) |
|
214
|
|
|
for i, lat in enumerate(latrange): |
|
215
|
|
|
if i % (len(latrange) / 10) == 0: |
|
216
|
|
|
logger.debug(str(int(i / len(latrange) * 100)) + '%') |
|
217
|
|
|
|
|
218
|
|
|
for j, lon in enumerate(lonrange): |
|
219
|
|
|
x, y, z = gps.lla2ecef([lat, lon, altitude]) |
|
220
|
|
|
distances, indexes = kdtree.query([x, y, z], n_nearest) |
|
221
|
|
|
min_travel_time = sys.float_info.max |
|
222
|
|
|
for distance, index in zip(distances, indexes): |
|
223
|
|
|
if stations[index].travel_time_min is None: |
|
224
|
|
|
continue |
|
225
|
|
|
travel_time = stations[index].travel_time_min + distance / 1000.0 / cycle_speed_kmh * 60.0 |
|
226
|
|
|
if travel_time < min_travel_time: |
|
227
|
|
|
min_travel_time = travel_time |
|
228
|
|
|
Z[i][j] = min_travel_time |
|
229
|
|
|
data = ContourData() |
|
230
|
|
|
data.index_begin = position |
|
231
|
|
|
data.Z = Z |
|
232
|
|
|
q.put(data) |
|
233
|
|
|
logger.info('end interpolate_travel_time') |
|
234
|
|
|
return |
|
235
|
|
|
|
|
236
|
|
|
|
|
237
|
|
|
# def contour_to_json(contour, filename, contour_labels, min_angle=2, ndigits=5): |
|
238
|
|
|
# # min_angle: only create a new line segment if the angle is larger than this angle, to compress output |
|
239
|
|
|
# collections = contour.collections |
|
240
|
|
|
# with open(filename, 'w') as fileout: |
|
241
|
|
|
# total_points = 0 |
|
242
|
|
|
# total_points_original = 0 |
|
243
|
|
|
# collections_json = [] |
|
244
|
|
|
# contour_index = 0 |
|
245
|
|
|
# assert len(contour_labels) == len(collections) |
|
246
|
|
|
# for collection in collections: |
|
247
|
|
|
# paths = collection.get_paths() |
|
248
|
|
|
# color = collection.get_edgecolor() |
|
249
|
|
|
# paths_json = [] |
|
250
|
|
|
# for path in paths: |
|
251
|
|
|
# v = path.vertices |
|
252
|
|
|
# x = [] |
|
253
|
|
|
# y = [] |
|
254
|
|
|
# v1 = v[1] - v[0] |
|
255
|
|
|
# x.append(round(v[0][0], ndigits)) |
|
256
|
|
|
# y.append(round(v[0][1], ndigits)) |
|
257
|
|
|
# for i in range(1, len(v) - 2): |
|
258
|
|
|
# v2 = v[i + 1] - v[i - 1] |
|
259
|
|
|
# diff_angle = math.fabs(angle(v1, v2) * 180.0 / math.pi) |
|
260
|
|
|
# if diff_angle > min_angle: |
|
261
|
|
|
# x.append(round(v[i][0], ndigits)) |
|
262
|
|
|
# y.append(round(v[i][1], ndigits)) |
|
263
|
|
|
# v1 = v[i] - v[i - 1] |
|
264
|
|
|
# x.append(round(v[-1][0], ndigits)) |
|
265
|
|
|
# y.append(round(v[-1][1], ndigits)) |
|
266
|
|
|
# total_points += len(x) |
|
267
|
|
|
# total_points_original += len(v) |
|
268
|
|
|
# |
|
269
|
|
|
# # x = v[:,0].tolist() |
|
270
|
|
|
# # y = v[:,1].tolist() |
|
271
|
|
|
# paths_json.append({u"x": x, u"y": y, u"linecolor": color[0].tolist(), u"label": str(int(contour_labels[contour_index])) + ' min'}) |
|
272
|
|
|
# contour_index += 1 |
|
273
|
|
|
# |
|
274
|
|
|
# if paths_json: |
|
275
|
|
|
# collections_json.append({u"paths": paths_json}) |
|
276
|
|
|
# collections_json_f = {} |
|
277
|
|
|
# collections_json_f[u"contours"] = collections_json |
|
278
|
|
|
# fileout.write(json.dumps(collections_json_f, sort_keys=True)) # indent=2) |
|
279
|
|
|
# logger.info('total points: ' + str(total_points) + ', compression: ' + str(int((1.0 - total_points / total_points_original) * 100)) + '%') |
|
280
|
|
|
|
bartromgens /
nsmaps
