How to determine Spatial Resolution in an image?

Question

I am new to the field of cameras and imaging. I am working with an industrial camera.

I need to check if the camera captured sufficient detail in the image and whether the images are good enough to be used for further analysis. What I came across while searching on the internet is spatial resolution.

After doing some experiments with image capture, what I learned is that if I capture the zoomed-in image, then spatial resolution increases. Is this correct? What factors determine this value? Is spatial resolution limited by the camera sensor and lens specifications? Or does it have to do with the way an image is captured? Will the spatial resolution change if I capture an image from an angle? Do focussing range and aperture affect the spatial resolution? Is spatial resolution the only thing that I should look for?

I am calculating the spatial resolution by getting the number of pixels covering 1mm of area in the image.

Answer 1

Optical capabilities: Well studied by Astronomer Royal John William Strutt, 3th Baron Rayleigh (Nobel Prize Physics). Valid today – The resolving power of a lens decreases as the aperture is decreased. Resolution decreases with increasing wavelength. It is about twice as great for extreme blue as per extreme red. The Rayleigh Criterion supplies the maximum theoretical resolving power in lines per millimeter. Resolving Power = 1392 ÷ f-number. The value 1392 is about the center of the photographic pictorial range of colors.

f/1.4 = 994 lines per mm

f/2 = 696 lines per mm

f/2.8 =487 lines per mm

f/4 = 348 lines per mm

f/5.6 = 249 lines per mm

f/8 = 174 lines per mm

f/11 = 127 lines per mm

f/16 = 87 lines per mm

This data is for on-axis objects, its worst for off-axis objects. This reduction in resolution as you stop down is due to the way light propagates. We are taking about the combined nemeses of diffraction and interference. Note that the resolution at f/8 exceeds most pictorial film ability.

To calculate the total resolution of a system, we test with parallel ruled test charts and insect to see how many lines per millimeter can actually be resolved. This will be a variable based on subject contrast and camera exposure and how well the lens is corrected for aberrations.

Answer 2

By chance, I've been on/off working on Spatial Resolution (albeit in a different kind of photography/context) during the last year.

It's not a hugely complicated subject itself, but it heavily depends on what are you really trying to achieve: if you are doing it for internal and/or personal use just invent one metric you feel is fine for your needs and, as long as you stick to it, you will be fine.

On the other hand, if you are doing this for some kind of specialized software that must give "official" results, your best bet is to stick to official image analysis documentation and procedures. There are worldwide standards and procedures, and at that level of requirements you want to use something that has been deeply studied and globally approved. And this in turn is the response to your other questions:

What factors determine this value? Is spatial resolution limited by the camera sensor and lens specifications? Or does it have to do with the way an image is captured? Will the spatial resolution change if I capture an image from an angle? Do focussing range and aperture affect the spatial resolution? Is spatial resolution the only thing that I should look for?

Yes, there are lot of factors weighting in and SR alone is never a good indicator of the quality of a picture for further analysis. You need to do lot of different kind of tests on the picture and it's ok only when all this different measures fall in the acceptable range.

If you find that you need to go in this direction, one of your first step could be to check with ASTM International and CEN.