-
Notifications
You must be signed in to change notification settings - Fork 2
/
create_animations.py
197 lines (165 loc) · 7.95 KB
/
create_animations.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
import os
import logging
import argparse
import sys
import csv
import subprocess
import rasterio
import imageio
import psutil
from numpy.ma import masked
from PIL import Image
from PIL import ImageDraw
from PIL import ImageFont
from moviepy.editor import VideoClip
from glob import glob
from future.utils import viewitems
from argparse import ArgumentParser
from collections import defaultdict
from collections import OrderedDict
from multiprocessing import Pool
from multiprocessing import cpu_count
class FrameGenerator(object):
def __init__(self, name, files, output_path, gdal_mem=64):
self.name = name
self.files = files
self.output_path = output_path
self.color_table = None
self.color_table_file = None
self.gdal_mem = gdal_mem
def scale(self, in_size, out_size):
in_width, in_height = in_size
out_width, out_height = out_size
scale_factor = float(out_width) / in_width if in_width > in_height else float(out_height) / in_height
return (int(in_width * scale_factor), int(in_height * scale_factor))
def make_frame(self, t):
if not self.color_table:
self.create_color_table()
in_file = self.files[int(t)]
frame_file = os.path.join(self.output_path, "{}_{}.png".format(self.name, int(t)))
self.colorize(in_file, frame_file)
# Add extra information to the colorized image: year, legend, etc.
img = Image.open(frame_file)
out_size = (1280, 720)
img = img.resize(self.scale(img.size, out_size))
legend_line_width = 2
legend_y_offset = 30
legend_top_label = str(round(max(self.color_table.keys()), 4))
legend_bottom_label = str(round(min(self.color_table.keys()), 4))
legend_label_width = max((len(legend_top_label), len(legend_bottom_label))) * 15
legend_size = (legend_label_width, len(self.color_table) * legend_line_width)
out_img = Image.new(img.mode,
(img.size[0] + legend_size[0], max(img.size[1], legend_size[1] + legend_y_offset * 2)),
color=(255, 255, 255, 255))
out_img.paste(img)
canvas_size = out_img.size
canvas = ImageDraw.Draw(out_img)
font = ImageFont.truetype("arial.ttf", 20)
year = os.path.splitext(in_file)[0].rsplit("_", 1)[1]
canvas.text((5, canvas_size[1] - legend_y_offset), year, font=font, fill=(0, 0, 0, 255))
canvas.text((5, 5), self.name, font=font, fill=(0, 0, 0, 255))
legend_start = (canvas_size[0] - legend_label_width, legend_y_offset)
legend_end = (legend_start[0], legend_size[1] + legend_start[1])
canvas.text((legend_start[0], 0), legend_top_label, font=font, fill=(0, 0, 0, 255))
canvas.text(legend_end, legend_bottom_label, font=font, fill=(0, 0, 0, 255))
for i, value in enumerate(sorted(self.color_table, reverse=True)):
color = self.color_table[value]
line_start = (legend_start[0] + legend_label_width / 2 - 10, legend_start[1] + i * legend_line_width)
line_end = (legend_start[0] + legend_label_width / 2 + 10, legend_start[1] + i * legend_line_width + legend_line_width)
canvas.rectangle((line_start, line_end), fill=color, outline=color)
out_img.save(frame_file)
data = imageio.imread(frame_file)
os.remove(frame_file)
os.remove("{}.aux.xml".format(frame_file))
return data
def create_color_table(self):
min_max_by_file = [self.get_min_max(file) for file in self.files]
min_value = min((stats[0] for stats in min_max_by_file))
max_value = max((stats[1] for stats in min_max_by_file))
step = float(max_value - min_value) / 256.0
self.color_table = OrderedDict((min_value + i * step, (125, i, 0, 255)) for i in range(255))
self.color_table_file = os.path.join(self.output_path, "{}_color.txt".format(self.name))
with open(self.color_table_file, "w") as color_file:
writer = csv.writer(color_file)
color_map = [["nv", 255, 255, 255, 255]] + [[val] + list(colors) for val, colors
in viewitems(self.color_table)]
writer.writerows(color_map)
def colorize(self, in_file, out_file):
subprocess.check_call([
"gdaldem",
"color-relief",
os.path.abspath(in_file),
os.path.abspath(self.color_table_file),
os.path.abspath(out_file),
"-of", "png",
"-q",
"--config", "GDAL_CACHEMAX", str(self.gdal_mem)])
def get_min_max(self, file):
try:
with rasterio.open(file, "r") as src:
raster_data = src.read(masked=True)
first_band = raster_data[:1]
min = first_band.min()
max = first_band.max()
return (min if min is not masked else 0,
max if max is not masked else 0)
except:
return (0, 0)
def create_animation(name, files, output_path, worker_mem):
'''
Creates an animation from a list of tiff files.
'''
sys.stdout = open(os.devnull, "w")
sys.stderr = open(os.devnull, "w")
# Add the last file twice to avoid it getting dropped from the animation.
files.extend(files[-1:])
frame_generator = FrameGenerator(name, files, output_path, worker_mem)
duration = len(files)
animation = VideoClip(frame_generator.make_frame, duration=duration)
animation.write_videofile(os.path.join(output_path, "{}.avi".format(name)),
fps=1, codec="png")
return name
def find_spatial_output(root_path):
'''
Gathers up all spatial output rooted in root_path by indicator and year.
'''
spatial_output = defaultdict(list)
for dir, subdirs, files in os.walk(root_path):
for file in sorted(filter(lambda f: f.endswith(".tiff"), files)):
file_path = os.path.abspath(os.path.join(dir, file))
indicator, year = os.path.splitext(file)[0].rsplit("_", 1)
spatial_output[indicator].append(file_path)
return spatial_output
def process_spatial_output(spatial_output, output_path=".", pool_size=1):
'''
Creates an animation for each indicator from a dictionary of
{indicator: {year: [.tiff files]}}
'''
if not os.path.exists(output_path):
os.makedirs(output_path)
available_mem = psutil.virtual_memory().available
worker_mem = int(available_mem * 0.8 / pool_size)
pool = Pool(pool_size)
for indicator, files in viewitems(spatial_output):
pool.apply_async(create_animation, (indicator, files, output_path, worker_mem),
callback=logging.info)
pool.close()
pool.join()
for palette_file in glob(os.path.join(output_path, "*_color.txt")):
os.remove(palette_file)
if __name__ == "__main__":
logging.basicConfig(stream=sys.stdout, level=logging.INFO)
parser = ArgumentParser(description="Creates an animation from a timeseries of geotiffs.")
parser.add_argument("--indicator_root", required=False, default=".",
help="path to the spatial output root directory")
parser.add_argument("--output_path", required=False, default=".",
help="path to store generated animation in - will be created if it doesn't exist")
parser.add_argument("--pool_size", help="Process pool size", required=False, default=1, type=int)
args = parser.parse_args()
spatial_output = find_spatial_output(args.indicator_root)
if spatial_output:
logging.info("Found {} total spatial indicators.".format(len(spatial_output)))
process_spatial_output(spatial_output, args.output_path, args.pool_size)
logging.info("Done")
else:
logging.info("No spatial output found.")