FL0.01 (approx)
To answer the camera related part first:
Since you had the lens "zoomed" it is unlikely that there was a fisheye effect due to the focal length of the lens. In fact, if anything one might expect the lens to lessen any apparent curvature at that setting.
If you are concerned about distortion in the lens there is a simple test, if you can be bothered. Find a conveniently large man made object which you know (or assume) is rectilinear - a tall office block is a useful one. From a fair distance back, such that it does not appear distorted to the naked eye, take a few photos of it. Rotate the camera through a few random angles and repeat. If the developed pictures look fairly consistent you can probably be reasonably confident that the lens is good. (It's hardly a very scientific method, but probably good enough)
Regarding the FL question, and my rather flippant answer above. In theory the curvature of the earth should be visible from any altitude - the problem is that irregularity in the earth's surface masks the effect at lower altitudes, and the observer introduces a perception issue.
Essentially, from any altitude the observer forms a cone, with his location at the tip and the horizon being the base of the cone. Looking along the surface of the cone - which is what you do when you look at the 'horizon', for that is simply the cone's surface, the base appears curved. The higher you go, the more pronounced the curve. But at any height above that of a worm, it's theoretically visible.
Oh, and to keep any flat-earthers and Discworld fans happy, you aren't seeing the curvature of the surface - just the endge of the surface. We know the earth is a sphere now, and interpret it accordingly. But a flat disc (with optional elephants and turtle) would look the same.
For the case you quoted, at that FL you'd have a visual range to the horizon of ~270nm. If you consider the cone I mentioned before, it has a height of about 18nm (the base of the cone runs well below the earth under you) and a radius of some 270nm also. A very shallow cone, you're not that high. If we consider your camera lens has about a 30 deg field of view, and was looking along the cone, then the circular base of the cone would pass out of view 15 degs either side of the centre of the picture, or about 70 nm each side of the centre of the picture. In 70 nm the earth curves by about 1 deg, and you have 1 deg either side. So you should have about 2 degs of curvature in your picture, which ought to be discernable.
Stirring things up is the possibility of irregularity of the horizon. You'd really need no significant variation in the altitude of the horizon (whether clouds, sea, whatever) to make sure that what you were seeing was really the curvature.