With one extra step, you could get a lower bound with very high confidence. Just...

		eirikbakke 6 months ago \| parent \| context \| favorite \| on: I counted all of the yurts in Mongolia using machi... With one extra step, you could get a lower bound with very high confidence. Just take a random sample (N=100) of the detected objects, and determine manually what percentage are really yurts. Then multiply that percentage by the number of detected objects.