Sensor Pitch and Resolving Power

Navigation: Pillar Guide | Wave Physics | Sensor Pitch and Resolving Power

Review status: pending. [Reviewer name and credential to be confirmed]. No reviewer has yet approved this page.

What pixel pitch means

Pixel pitch is the distance from the centre of one pixel to the centre of the next. A sensor with a pitch of 4 micrometres records detail in steps of that size. Smaller pitch means more pixels in the same area, but each pixel collects less light, which affects noise.

The Nyquist limit

To record a repeating pattern without distortion, a sensor must sample it at least twice per cycle. The highest spatial frequency that can be recorded faithfully is the Nyquist frequency, equal to 1 divided by twice the pixel pitch. For a 4 micrometre pitch, that is 1 divided by 0.008 millimetres, which is 125 line pairs per millimetre.

Detail finer than the Nyquist frequency is not recorded correctly. It appears as false patterns, called aliasing, often seen as moiré in fabric or brickwork.

Lens and sensor together

A lens produces a detail limit set by diffraction and aberrations, and a sensor produces a second limit set by its pixel pitch. The overall detail in the image is limited by whichever is more restrictive. A sensor finer than the lens can resolve adds pixels without adding detail, while a lens sharper than the sensor sends more detail than the sensor can record.

Anti-aliasing filters

Many cameras use an optical low-pass filter in front of the sensor. It blurs fine detail slightly so that detail beyond the Nyquist frequency is less likely to alias. The trade-off is a small loss of sharpness at all frequencies. Some cameras omit the filter, accepting more aliasing risk for more detail.

Common mistakes

Treating pixel count as resolution. Comparing pixel pitch between sensors of different sizes without considering the image size. Assuming aliasing can be fixed in post-processing without any loss.

Sensor size and pixel pitch together

Pixel pitch alone does not describe resolution. Two sensors with the same pitch but different sizes produce different image dimensions, and the lens required to cover each sensor differs. The useful comparison is detail per unit of image size, for example line pairs per millimetre on the sensor, or the number of pixels across the frame at a given field of view.

Practical checks

Photograph a fine test pattern near the Nyquist frequency of the sensor, and look for false colours or moiré. If they appear, the sensor or the lens is delivering detail beyond what the pixels can record. Changing the aperture or using a different anti-aliasing setting can change how much aliasing appears, which helps identify the cause.

Balancing pixel size against noise

Smaller pixels give finer sampling, but each pixel receives fewer photons in a given exposure, which increases noise in dim conditions. Larger pixels collect more light and reduce noise, at the cost of coarser sampling. The best pixel pitch depends on whether the priority is fine detail or low noise, and on the light the camera is expected to work in.

Microlenses and fill factor

Most sensors include a microlens over each pixel to direct light onto the light-sensitive area. The fill factor is the fraction of the pixel area that is light-sensitive. Microlenses help raise the effective light gathering, but at the very smallest pixel sizes, the diffraction limit of the microlens itself can matter, and this affects the angle at which light is accepted.

Sensor and lens interplay at the edge

The angle at which light reaches the edge of the sensor is usually steeper than at the centre. Some sensors are designed for a specific angle, and lenses designed for other sensors may show colour shifts or reduced sharpness at the edges. The chief ray angle matters for matching lenses and sensors, and it is one reason some lenses are described as suitable for particular cameras.

Reading the image for aliasing

Aliasing shows as false patterns or colour bands in repeating detail that lies near or above the Nyquist frequency. Look at fabrics, screens or printed grids, and compare the result with a higher-resolution capture. If the false detail disappears at a longer focal length, the lens may be delivering finer detail than the sensor can sample at that magnification.

Summary of the sensor limit

The sensor limit is set by pixel pitch and the Nyquist frequency, and it is independent of the lens. When the lens can deliver detail finer than the Nyquist frequency, aliasing risk rises unless an anti-aliasing filter is used. When the lens delivers less detail than the sensor can sample, the sensor adds pixels without adding information. The better of the two is set by the combination.

Choosing a sensor for detail

When detail is the priority, compare the pixel pitch with the lens diffraction limit at the apertures you use. A sensor much finer than the lens limit adds files without adding detail. A sensor coarser than the lens limit can use the lens fully, but may alias fine repeating patterns.

Testing on your own equipment

Use a high-contrast test chart with known line spacing, photographed at several apertures. Measure the frequency where contrast falls sharply, and compare it with the Nyquist frequency of your sensor. The difference shows whether the lens or the sensor is the limit for your setup.

Going further

The sensor limit is easy to overlook because it is invisible in the lens data. Checking the pixel pitch against the Nyquist frequency is a simple habit that prevents many disappointing upgrades.

Summary

The key point of this article is the one stated at its start. Check the formula, use the stated assumptions, and confirm any result with a test before relying on it in practice.

Sources

Eugene Hecht, Optics, Pearson (any current edition).
Max Born and Emil Wolf, Principles of Optics, Cambridge University Press (any current edition).
ISO 12233: Photography, Electronic still picture imaging, Resolution and spatial frequency responses.

Next step

Return to the pillar guide or the Wave Physics summary to read the related articles.

Who should review this: This technical documentation should be reviewed by a credentialed expert. Reviewer: [name and credential to be confirmed]. Last reviewed: [date to be added at review].

General information only. This page explains optical principles and does not replace lens design work, lens testing or advice from a qualified optical engineer.