You couldn’t tell the difference between a single 6-inch line vs. two closely-spaced 3-inch lines. However, road lines are all approximately the same width, so double lines have more effective surface area and would appear brighter than single lines. You could even determine passing areas from the alternating brightness, though you wouldn’t know which side of the road had the right.
Unless the double line were more faded/worn than the single line.
Wouldn’t something under the size of a single film grain or imaging element more or less average out with the other stuff covered by that element/grain?
In other words, if your spy satellite was looking at a big area of this kind of tile with 1" hexagons, wouldn’t you end up with white and gray image elements, as the black ones would be averaged into the surrounding white ones?
My dad was an Air Force photo interpreter, who mostly worked with aircraft photos in the Viet Nam era. He said that the secret to being a good PI was being able to see the story of what was going on, not just seeing what was there.
‘What’ is in the photo is only the first question. You follow up with ‘why’, and then apply a lot of if/then.
For a building, in addition to the staff, you can also look at it’s size and shape, windows and doors, security features (fences and guard stations), transportation services (rail sidings or just road, what kind of road) and even what kind of roof (along with roof top facilities). You would also consider the neighboring buildings, as well as the surrounding area in general.
The saying is ‘a picture is worth 1000 words’, but any one who says that never met a PI.
Yes, which is why feature size and contrast are important. If your features are smaller than the resolution, then contrast will be decreased. If your contrast was good enough to begin with, then even the decreased contrast will be enough. If your initial contrast wasn’t so good, though, or if the feature size was enough smaller, then the contrast will end up too low to distinguish.
A lot of it is cultural as well. Good photo interpretation requires you to be able almost “feel” what’s going on. It’s an art form. In the US, most of our schools are set up with baseball and football fields, and tracks, our shopping malls may look different. Our agriculture, suburban developments, universities, community centers, factories all have some cultural elements in their design that may or may not differ from a different country.
This articlemade me think of this thread.
Seems to me that a government with enough resources could feed all their recon photos into one of those computer systems and teach it how to recognize things, and set it loose from there to identify things.
Tell that to this lady. http://trekfoundations.files.wordpress.com/2014/12/illia.jpg 
So was the Russian babe in the woods a natural blonde or not?