How do they keep from double counting during a bird count?

Actually, they count the number of wings. Then they divide by two.

That doesn’t sound practical, unless they use a photograph. And counting wings? Oh my!

If you do a bird count in your garden, or veranda, or what have you, you count for half an hour (or any given length of time ordered by a bird counting authority).
If you see at one point three blue tits in a tree, and 10 minutes later you see five, you only count the five (or the highest number of one species witnessed in that half hour).
If you count migrating birds it is a good idea to make photographs of flocks and count them afterwards (obviously that’s easier than counting them in flight).

When birds are count in those special bird counting events the numbers that come out are rough estimates, I presume, so a certain percentage of double counted species is probably calculated beforehand.

I’ve wondered this same thing.
How do they handle this ?

To compensate for this possibility, they probably only consider a percentage of the total count.

I won’t even attempt to explain how they do it especially when the numbers grow. Read here to learn about the various techniques used. I think the one that fascinated me the most was when a large flock is flying overhead!!! Takes me back to the old saying where there’s a will, there’s a way.