108 lines
3.0 KiB
Python
108 lines
3.0 KiB
Python
|
#!/usr/bin/env python
|
||
|
|
|
||
|
|
from PIL import Image
|
||
|
|
import colorsys
|
||
|
|
import json
|
||
|
|
import math
|
||
|
|
import optparse
|
||
|
|
import os
|
||
|
|
import re
|
||
|
|
|
||
|
|
|
||
|
|
class Space:
|
||
|
|
def __init__(self):
|
||
|
|
self.data = dict()
|
||
|
|
|
||
|
|
def load(self, filename):
|
||
|
|
self.data = dict()
|
||
|
|
self.data['features'] = features = list()
|
||
|
|
self.data['keywords'] = keywords = dict()
|
||
|
|
|
||
|
|
with open(filename, 'r') as fp:
|
||
|
|
space = json.load(fp)
|
||
|
|
|
||
|
|
colors = space['colorString']
|
||
|
|
impressions = space['impressions']
|
||
|
|
|
||
|
|
for color in colors:
|
||
|
|
groups = re.split('[\(\)]', color)
|
||
|
|
name = groups[0]
|
||
|
|
rgb = map(lambda x: int(x), groups[1].split(','))
|
||
|
|
features.append({
|
||
|
|
'name': name,
|
||
|
|
'rgb': rgb,
|
||
|
|
'hsv': colorsys.rgb_to_hsv(*rgb)
|
||
|
|
})
|
||
|
|
|
||
|
|
for impression in impressions:
|
||
|
|
assert len(colors) == len(impression) - 1
|
||
|
|
keywords[impression[0]] = impression[1:]
|
||
|
|
|
||
|
|
def save(self, filename):
|
||
|
|
with open(filename, 'w') as fp:
|
||
|
|
fp.write('var spaceDefinitions = ')
|
||
|
|
json.dump(self.data, fp, indent=4, sort_keys=True)
|
||
|
|
|
||
|
|
|
||
|
|
class Sensor:
|
||
|
|
def __init__(self):
|
||
|
|
self.space = Space()
|
||
|
|
self.rankings = dict()
|
||
|
|
|
||
|
|
def prepare_definitions(self, filename_in, filename_out):
|
||
|
|
self.space.load(filename_in)
|
||
|
|
self.space.save(filename_out)
|
||
|
|
|
||
|
|
def process_images(self, filenames):
|
||
|
|
for filename in filenames:
|
||
|
|
self.process_image(filename)
|
||
|
|
|
||
|
|
def process_image(self, filename):
|
||
|
|
print 'Processing {0}...'.format(filename)
|
||
|
|
|
||
|
|
image = Image.open(filename)
|
||
|
|
width, height = image.size
|
||
|
|
rankings = list()
|
||
|
|
|
||
|
|
for feature in self.space.data['features']:
|
||
|
|
feature_rgb = feature['rgb']
|
||
|
|
print '\t{name}...'.format(**feature)
|
||
|
|
|
||
|
|
difference = 0
|
||
|
|
for y in xrange(height):
|
||
|
|
for x in xrange(width):
|
||
|
|
difference += self.color_distance(image.getpixel((x, y)), feature_rgb)
|
||
|
|
rankings.append(difference / (width * height))
|
||
|
|
|
||
|
|
self.rankings[os.path.basename(filename)] = rankings
|
||
|
|
|
||
|
|
|
||
|
|
def color_distance(self, color1, color2):
|
||
|
|
dx = color2[0] - color1[0]
|
||
|
|
dy = color2[1] - color1[1]
|
||
|
|
dz = color2[2] - color1[2]
|
||
|
|
|
||
|
|
return dx * dx + dy * dy + dz * dz
|
||
|
|
|
||
|
|
def save_database(self, filename):
|
||
|
|
with open(filename, 'w') as fp:
|
||
|
|
fp.write('var spaceDatabase = ')
|
||
|
|
json.dump(self.rankings, fp, indent=4, sort_keys=True)
|
||
|
|
|
||
|
|
|
||
|
|
def main():
|
||
|
|
parser = optparse.OptionParser()
|
||
|
|
parser.add_option('--space', dest='space', default='space.json')
|
||
|
|
parser.add_option('--definitions', dest='definitions', default='definitions.json')
|
||
|
|
parser.add_option('--database', dest='database', default='database.json')
|
||
|
|
options, filenames = parser.parse_args()
|
||
|
|
|
||
|
|
sensor = Sensor()
|
||
|
|
sensor.prepare_definitions(options.space, options.definitions)
|
||
|
|
sensor.process_images(filenames)
|
||
|
|
sensor.save_database(options.database)
|
||
|
|
|
||
|
|
|
||
|
|
if __name__ == '__main__':
|
||
|
|
main()
|