Spatial resolution, for an infrared camera, is based on detector pixels and the field of view specification, combining them to define the area the camera sees at any given moment.
It is the least glamorous specification on a thermal imaging camera and the one that most often decides whether a measurement is true.
The three numbers
Field of view (FOV) is the total angle the lens takes in — for example 24° × 18°. At a given distance that becomes a rectangle of a certain size on the target.
Instantaneous field of view (IFOV) is the angle subtended by a single detector pixel. Divide the FOV by the detector resolution and you have it. It is usually quoted in milliradians (mrad), and it tells you the smallest detail the camera can see.
Measurement field of view (MFOV) is the smallest area from which the camera can return an accurate temperature. It is not the same as IFOV, and this is the distinction that matters.
Why seeing and measuring are different
A single pixel will register that something is there. It will not necessarily register how hot it is, because energy from adjacent areas bleeds into the reading and the optics are not perfect.
The working rule is that the target needs to occupy roughly three by three pixels for a trustworthy temperature. So the measurement spot is about three times the linear size of the detail the camera can resolve.
In practice this means you can see a hot bolt in an image at a distance where you cannot measure it.
Working it out in the field
The arithmetic is simple and worth doing:
Spot size = IFOV (in mrad) × distance (in metres)
A camera with an IFOV of 1.3 mrad at 3 metres resolves about 3.9 mm per pixel. Applying the 3× rule, the smallest reliably measurable target is around 12 mm.
That is fine for a busbar or a cable lug. It is marginal for a small terminal screw, and it is useless for a strand of a conductor.
Double the distance to 6 metres and the measurable target becomes about 24 mm. Distance is expensive, and it is why survey positions matter as much as camera specification.
Why the error always runs the same way
When a hot target is smaller than the measurement spot, the camera averages the target with everything around it — which is cooler.
The reported temperature is therefore always too low. Never too high.
This is what makes the error dangerous rather than merely annoying. A genuine Priority 1 fault, measured from too far away with too wide a lens, can present as a mild anomaly or as nothing at all. Every other significant source of error in electrical thermography — emissivity, reflection, load — fails in the same direction.
What to do about it
- Get closer. The cheapest fix, where access and safety permit.
- Fit a telephoto lens. A 12° lens roughly halves the spot size against a 24° lens at the same distance.
- Use a higher-resolution detector. More pixels across the same field of view means a smaller IFOV.
- Fill the frame with the target wherever the finding is small and hot.
- Note the limitation in the report where none of the above is possible. A measurement you cannot trust should be described as such rather than quoted as a number.
That last point is the professional one. A report that quotes a confident temperature on a target smaller than the spot size is worse than a report that says the measurement could not be made reliably.
Where it fits with everything else
Spatial resolution is one of three surface and geometry effects that determine whether a thermographic measurement is defensible. The other two — emissivity and reflection — are covered in diagnosing electrical problems safely with thermal imaging.
Together they are the reason certification matters. Any operator can produce an image. Knowing which numbers in that image can be relied upon is a Category 2 function under ISO 18436-7.