How do you calculate camera resolution in mm per pixel?

Divide the horizontal field of view by the horizontal pixel count. A 220 mm FOV across a 2448-pixel sensor gives 220 / 2448 = 0.090 mm per pixel.

What is a good resolution for machine vision inspection?

A good rule is to have at least 3-5 pixels across the smallest feature you must detect. At 0.090 mm per pixel a defect needs to be roughly 0.27-0.45 mm before it is reliably visible.

Is more megapixels always better for inspection?

Not always. More pixels improve resolution only if the lens can resolve them and the FOV stays fixed; otherwise you pay for data and processing time you cannot use. Match pixels to the tolerance, not to the spec sheet.

What does the unit conversion factor do here?

It rescales the result into your preferred units, for example using 1000 to express microns per pixel instead of mm per pixel. Left at 1, the output stays in mm per pixel as in the 0.090 example.

Spatial resolution vs measurement accuracy?

Spatial resolution is the pixel size at the part; accuracy is how precisely you can locate an edge, often a fraction of a pixel with subpixel algorithms. Resolution sets the floor, but accuracy can be better or worse depending on contrast and calibration.

How do I improve resolution without a new camera?

Shrink the FOV by using a longer lens or moving closer, which puts the same pixels across a smaller part. Cutting FOV from 220 mm to 110 mm halves resolution to about 0.045 mm per pixel.

Machine Vision & Industrial Inspection AI calculator

Camera Resolution Calculator

Spatial resolution in machine vision is how many millimeters of the real part each camera pixel covers, expressed as mm per pixel. It is the single number that links your optics to your inspection capability: a fine resolution lets the algorithm see small features, a coarse one blurs them away. Integrators and quality engineers compute it to confirm a camera can actually measure the tolerances and defects the application demands before any code is written. This calculator divides the horizontal field of view by the horizontal pixel count to give resolution at the part.

What this calculator does

Calculate inspection resolution (mm per pixel) for a machine vision camera by dividing the horizontal field of view by the camera horizontal pixel count, so you can confirm whether the camera has enough resolution to meet your inspection specification.
Use it when specifying a machine vision system and you need to confirm that the chosen camera and lens combination gives enough mm per pixel resolution to reliably detect the smallest defect in your inspection specification.
It computes inspection spatial resolution in mm per pixel by dividing the horizontal field of view by the camera's horizontal pixel count.

Formula used

Inspection resolution = FOV width / horizontal pixel count x conversion factor
Example: 220 mm / 2448 pixels x 1 = 0.090 mm per pixel

Inputs explained

Horizontal field of view (FOV width):
Camera horizontal pixel count:
Unit conversion factor:

How to use the result

Use it after fixing the FOV to verify the camera has enough pixels to meet your measurement tolerance or defect-size requirement.
It reports geometric pixel pitch only; real achievable resolution is worse because of lens MTF, defocus, motion blur and the need for several pixels per feature.

Common questions

How do you calculate camera resolution in mm per pixel? Divide the horizontal field of view by the horizontal pixel count. A 220 mm FOV across a 2448-pixel sensor gives 220 / 2448 = 0.090 mm per pixel.
What is a good resolution for machine vision inspection? A good rule is to have at least 3-5 pixels across the smallest feature you must detect. At 0.090 mm per pixel a defect needs to be roughly 0.27-0.45 mm before it is reliably visible.
Is more megapixels always better for inspection? Not always. More pixels improve resolution only if the lens can resolve them and the FOV stays fixed; otherwise you pay for data and processing time you cannot use. Match pixels to the tolerance, not to the spec sheet.
What does the unit conversion factor do here? It rescales the result into your preferred units, for example using 1000 to express microns per pixel instead of mm per pixel. Left at 1, the output stays in mm per pixel as in the 0.090 example.
Spatial resolution vs measurement accuracy? Spatial resolution is the pixel size at the part; accuracy is how precisely you can locate an edge, often a fraction of a pixel with subpixel algorithms. Resolution sets the floor, but accuracy can be better or worse depending on contrast and calibration.

Last reviewed 2026-05-12.