Automatically extracting bomb drops from the Blitz Bomb Census maps

I just finished Blackout by Connie Willis and I was wondering where she did get all the exact times when the air raids of The Blitz were happening.
I am curious because I was thinking that a timeline map would give a visual effect of the huge destruction and hard times the Londoners had to go through...so I googled for it and found the following:

• a nice research blog which uses the Blitz Bomb Census from The National Archives
• a blog entry from the Guardian about the first day - and data - of the Blitz over London.
   Here the source of info is the London Metropolitan Archives

Both examples show nice data visualization opportunities, but they both show how difficult it is to obtain data. Digital access to historical data could triggers a blossoming of applications!

Anyway I experimented with the little map from the first blog and wrote a little program using SimpleCV.

This is were I started from:

and this is what I got (without too much work, I must admit):

The code is quite simple and the parameters could be changed to get higher accuracy.

blz = Image("./sampleimages/the_blitz.png")
img = blz.copy()
dist = img.colorDistance(SimpleCV.Color.WHITE)
dist.dilate(2)

segmented = dist.stretch(240, 255)
blobs = segmented.findBlobs()
if blobs:
 circles = blobs.filter([b.isCircle(0.6) for b in blobs])
 for c in circles:
  img.drawCircle((c.x, c.y),
                 c.radius(),
                 SimpleCV.Color.YELLOW,
                 2)
 img.show()

wiki of the future?

Last weekend I have been watching, reading and playing around with Ward Cunningham's Smallest Federated Wiki. As usual he is a great designer and aims to simple, effective, understandable and useful tools! I particularly like the plugins idea (not new of course) and how easy it seems to be to add new ones, like support for MathJax:
Get inspired you too!