scrape-maps.py

import urllib
from PIL import Image
import cStringIO as StringIO
import numpy as np
import time

# index for naming images
n = 0

# iterate over the x/y locations of the map tiles
for x in range(2585, 2685, 2):
	print("Starting row: " + str(x))

	for y in range(6136, 6236, 2):

		try:
			# accumulator array for 2x2 tiles
			bigimg = np.zeros((512, 512, 3), np.uint8)

			for dx in range(2):
				for dy in range(2):
					# url format for scraping planet labs images
					url = "http://pm4.planet-labs.com/open_california_re_20130901_20131130/gmap/14/" + str(x + dx) + "/" + str(y + dy) + ".png"

					data = urllib.urlopen(url).read()

					stream = StringIO.StringIO(data)
					img = np.asarray(Image.open(stream))

					# place in 2x2 tile array
					bigimg[dy*256:(dy+1)*256, dx*256:(dx+1)*256] = img

			# save 2x2 tile image
			img = Image.fromarray(bigimg)
			img.save(str(n) + ".png", "PNG")

			# increment label iterator
			n = n + 1

			# let the server have a tiny little break
			time.sleep(0.01)

		except ValueError:
			print "ValueError, map tile unexpected size."
		except IOError:
			print "IOError, map tile missing."