Camera Sensor Size Guide

How Does CMOS Sensor Size Impact Camera Performance?

Sensor format sets the light-collecting area, and it drives physical size, cost, and lens requirements. It does not by itself set low-light SNR or dynamic range. Those come out of quantum efficiency, fill factor, full-well capacity, read noise, dark current, the exposure time and f-number you run, and the processing behind the sensor. Format is one input to that budget. Every camera project weighs it against the rest.

Larger pixels collect more photons per pixel, but that becomes better low-light performance only under a fair comparison: same f-number, same exposure time, same scene irradiance, comparable quantum efficiency and pixel architecture, and the same output resolution after any binning or downsampling. Compare two sensors at matched output resolution and the per-pixel advantage often shrinks.

A larger sensor also needs a lens with a larger image circle, and holding image quality and corner illumination across that wider field is the harder design problem. M12 lenses cover sensors up to 1/1.8 inch typically, with select designs to 2/3 inch, while C-mount lenses carry the formats above that.

Why Are Digital Camera Sensor Format Types So Confusing?

The format type classification system originates from video camera tubes manufactured before CCD and CMOS sensors. These vidicon tubes had opaque regions outside the active cathode area due to mechanical structures holding the tube and electrodes.

The imaging area was significantly smaller than the tube's outer diameter. When solid-state CCD sensors replaced tubes, manufacturers maintained the existing format naming convention to preserve compatibility with established lens systems. The result: modern sensor format designations describe the equivalent vidicon tube diameter that would produce the same image circle, not the sensor's actual dimensions.

What Is the Mathematical Relationship Between Format Type and Actual Sensor Size?

We've done our best to fit a "modern-day" curve to the format names, using commonly agreed upon datapoints of 1" = 16.0mm, 1/2" = 8.0mm, 1/3" = 6.0mm, 1/4" = 4.5mm, then cross-referencing as many other articles as possible. Treat that fit as a convenience, not a definition: the format names are a historical list inherited from tube diameters, not a continuous function of sensor diagonal.

That is why the curve breaks between the 1/2" and 1/2.3" format sizes, and why the chart below still carries the legacy "image dissector" label for the tube diameter. When the number matters, take width and height from the sensor datasheet or from the reference table below.

Sensor Format Reference Table

The table below lists common CMOS sensor formats with nominal dimensions. Active area varies between manufacturers and between parts sold under one label. Use the table for orientation and the datasheet active area for any calculation.

How Do I Match a Lens to My Sensor Format?

The lens image circle must equal or exceed the sensor diagonal. A full-frame fisheye lens is no exception: it is specified to cover the diagonal, and accepting barrel distortion does not substitute for coverage. A circular fisheye is the deliberate exception, since its image circle is designed to fall inside the sensor's short side.

A "1/2 inch" rating is shorthand for an image circle near 8mm, not a guarantee of coverage. Coverage is a criterion: ask what relative illumination and MTF the vendor still holds at the edge of that circle, and at which aperture, wavelength, and conjugate. Two lenses carrying the same format label can differ at the corner once the criterion is fixed. Run a sensor past the rated circle and the corners darken or go black.

Running a lens on a sensor smaller than its rated format is common and usually safe optically, but coverage alone is not compatibility. Chief ray angle, the sensor cover glass and filter stack, working distance, MTF over the field you actually use, mechanical clearance behind the mount, and the narrower field of view all still have to check out.

For embedded vision applications with sensors up to 1/1.8" format (8.9mm diagonal), M12 mount lenses provide compact, cost-effective solutions. For larger sensor formats, C-mount lenses extend coverage beyond what most M12 optics reach, and many C-mount models add an adjustable iris and focus ring for depth-of-field control.

What Sensor Format Should I Choose for My Application?

Sensor format selection depends on resolution requirements, light sensitivity needs, physical space constraints, and budget. Consider these guidelines:

• Mobile Robotics (1/4" to 1/1.7"): Compact sensors pair well with lightweight M12 lenses for robotics. Prioritize wide field of view for navigation.
• Automotive Vision (1/2.8" to 1/1.7"): Balance between compact packaging and low-light performance. Automotive M12 lenses offer ruggedized designs.
• Machine Vision (1/1.7" to 1"): Larger formats provide more area for a given pixel pitch. Global shutter is a pixel and readout architecture choice, available in both small and large sensors. Global shutter sensors in 2/3" and 1" formats pair with C-mount lenses for precision applications.
• Surveillance (1/3" to 1/2"): Cost-effective formats with proven lens availability. Surveillance lenses optimize for day/night operation.

How Does Exposure Interact With F-Number, Gain, and Motion Blur?

Sensor format sets the physical area available to collect light, but how much light actually reaches each pixel in a given frame comes down to exposure: image-plane irradiance multiplied by integration time. Aperture, transmission, and spectrum act through the irradiance term; sensor gain sits outside exposure, a readout setting that changes no photons. Image-plane irradiance scales with 1/(f/#)², so each full stop increase in f-number (for example F/2.8 to F/4) roughly halves the light reaching the sensor. See the f-number guide for the full derivation.

Lengthening exposure time is one way to recover light lost to a smaller aperture, but it also lengthens the smear from any object or camera motion during the frame. At typical conveyor speeds of 200–500mm/s, an extra 1ms of exposure time adds roughly 0.2–0.5mm of motion blur at the object. Our motion blur guide and camera exposure guide cover how this tradeoff affects detection accuracy in vision systems.

Sensor gain, the third adjustment, acts at readout, not on exposure. Analog gain, ahead of the analog-to-digital converter, suppresses only the noise added after the gain stage: ADC and quantization noise. Read noise accumulates across the reset, source-follower, column-amplifier, and ADC stages; noise already at the gain input is amplified with the signal, not suppressed. No gain setting collects photons the optics never delivered, and analog gain costs headroom at the bright end. Digital gain rescales values that were already converted, so it raises brightness and visible image noise together.

Because full-well capacity scales with pixel area, though full-well density is sensor-specific, sensors with larger pixels, which for a given resolution come from larger formats, hold more signal headroom before gain-driven noise becomes limiting. This is one more reason format selection is a system-level tradeoff, not just a field-of-view decision.

Frequently Asked Questions

Why doesn't a 1/2.8 inch sensor measure 1/2.8 inches?

The inch designation originates from 1950s vidicon television camera tubes. The format name referred to the outer glass tube diameter, not the active imaging area. The active cathode area was only about two-thirds of the tube diameter due to mechanical mounting structures.

When solid-state sensors replaced tubes, manufacturers retained the naming convention for backward compatibility with existing lens systems. A "1/2.8 inch" sensor has approximately 6.4mm diagonal, nowhere close to 0.357 inches (9.07mm).

How do I calculate field of view from sensor format?

Do not use format type directly for field of view calculations. The format designation is an approximation that varies between manufacturers. Instead, use the actual sensor dimensions from the datasheet.

Angle of view needs both the sensor dimension and the focal length: AoV = 2 × arctan(sensor_dimension / (2 × focal_length)), for a rectilinear lens focused near infinity. Linear field of view then needs a working distance: FOV = 2 × WD × tan(AoV / 2). Both are pinhole approximations. At finite conjugates the image distance exceeds the focal length, so work from magnification instead. Use our Field of View Calculator with actual sensor width or height in millimeters.

What happens if I use a lens rated for a smaller sensor format?

The lens image circle will not cover the entire sensor, causing vignetting (dark corners). The severity depends on the format mismatch. A 1/3" lens on a 1/2" sensor may show significant corner darkening. A 1/2" lens on a 2/3" sensor will have severe vignetting.

Always select a lens rated for your sensor format or larger. A 2/3" lens on a 1/3" sensor uses only the central portion of the image circle, which is optically safe, but coverage alone is not compatibility: check chief ray angle, the cover glass stack, rear clearance, and the narrower field of view before calling the pairing good.

How do I calculate the required focal length for my application?

Start with the far-conjugate approximation: Focal Length ≈ (Working Distance × Sensor Width) / Scene Width. It treats the lens as a pinhole, which is close enough when the object sits many focal lengths away.

Imaging a 2-meter wide scene from 3 meters with a 1/2.8" sensor (5.1mm width) gives 3000mm × 5.1mm / 2000mm = 7.65mm, so a 7.5mm or 8mm lens. At short working distances use the thin-lens form f = WD × m / (1 + m), with magnification m = sensor width / scene width and WD measured from the object-side principal plane. Use our EFL Calculator to check both.

How does f-number affect exposure and motion blur?

F-number controls how much light reaches the sensor per unit time: image-plane irradiance scales with 1/(f/#)², so a wider aperture (lower f-number) reaches the same exposure level in less time.

Shorter exposure time reduces motion blur, which scales roughly with object speed multiplied by exposure time. A narrower aperture forces a choice between a longer exposure (more blur), brighter illumination, or higher sensor gain (more noise, less dynamic range).

Can Commonlands assemble complete camera modules?

Yes. Commonlands provides camera module assembly services including optical design, sensor integration (Sony, OmniVision, OnSemi), lens holder selection, and complete module assembly with focusing and thread locking.

We support volumes from 100 to 100,000+ units annually. Contact our engineering team with your specifications.