Archive for March, 2007

Foreground Removal with Photoshop

Wednesday, March 21st, 2007

Stone Wall original

While walking in Duke Forest, I took a series of photos of this stone wall from slightly different vantage points. The idea was that I could then overlay the images in Photoshop and keep the good parts of each one, removing the foreground small trees and branches.

Stone Wall cleaned

It worked OK once I got the hang of layers, but there were still a couple of problems. Some of the vantage points were too different — I needed to be more careful about keeping the angles parallel. The other problem was that some rock parts were covered in all of the photos, so I had to fill in that area from other parts of the wall.

Bill tells me that’s what I should have done in the first place — just take one photo and use fake/duplicated rocks to replace the foreground material. My way seems more honest, even though the cleaned up image is still a fake.

Cape Fear MABC Regional 2007

Thursday, March 1st, 2007

My mother and I played bridge for three days solid at the Cape Fear MABC Regional in Wilmington last week. Actually, we only played about 11 hours per day, skipping the midnight sessions. We hoped to play a fourth day but got knocked out of a single elimination event in the second round and decided not to start a new event on the fourth day.

On the bright side, we won our bracket in a two-day compact knock-out event when we teamed up with a pair we’d never met before from Cary (Thanks Maryann and Dianne), which gave us a couple of coveted gold masterpoints. Plus we won a pairs event, and picked up some fractional points for other finishes.

The tournament attendance summary shows 1188 players playing 2336 session-tables, which means with 4 players per table per session that each player played on average about 8 sessions (2336 * 4 / 1188).

It’s interesting to note that 1118 (94%) players earned masterpoints at the event, suggesting performance is more random than expected. Masterpoints are generally awarded to the top third of the finishers of an event, and my crude conclusion is that anyone who plays about 8 sessions will likely get some masterpoints.
Looking a little further (but not far enough for much confidence), it would seem that placing is mostly luck. That is, if it were all luck, you would have a 1/3 chance of getting masterpoints for a single event, and the chance of getting some masterpoints over 8 events would be (1-(2/3))^8 = 96%. However, it more than just a matter of getting masterpoints. One might argue that skill is more of a factor in placing in the top 10% but not so much for placing in the top 33%.