SWIR Camera Lens for InGaAs Cameras: How to Match Sensor Size, Pixel Pitch, 2 MP Resolution and Optical Performance

Selecting a SWIR camera lens for an InGaAs camera is not simply a matter of choosing the correct mount and a focal length that produces the desired field of view. In a properly engineered 900–1700 nm machine-vision system, the lens, sensor dimensions, pixel pitch, image format, optical resolution, working distance and required defect size must operate as one matched imaging chain. A lens can physically attach to a camera yet still be poorly matched if its image circle does not cover the active sensor, its optical detail does not support the camera's pixel sampling, its focal length produces insufficient pixels on the smallest target, or its performance degrades in the spectral region that creates the actual inspection contrast. This is why InGaAs camera integration should begin with sensor-level calculations rather than choosing optics from focal length alone.

The dedicated Kyptec Automation® SWIR Camera Lens collection currently provides five focal lengths—8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm—for industrial SWIR imaging. The live portfolio identifies the lenses around 900–1700 nm wavelength coverage, 2 MegaPixel resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount, providing OEMs with a consistent optical family that can be evaluated against the geometry and sampling requirements of an InGaAs camera. The important engineering task is to understand what those specifications mean when connected to an actual sensor and application.

Why InGaAs Cameras and SWIR Lenses Must Be Selected Together

InGaAs imaging sensors are widely associated with short-wave infrared imaging because the detector material can respond effectively through wavelength regions beyond conventional silicon-based visible imaging. For an OEM, however, sensor sensitivity alone does not determine image quality. The lens must deliver the required field of view and sufficient optical contrast to the active sensor area while transmitting the wavelengths where the material difference is being inspected. The camera determines how the image is sampled; the lens determines what optical information reaches those samples. If either side of this relationship is poorly matched, the final machine can underperform even when both components appear strong individually.

This is especially important for material identification, moisture-sensitive inspection, silicon analysis, contamination detection and other SWIR applications in which small differences in intensity or spatial structure may determine the production decision. A high-quality InGaAs camera cannot recover detail that the lens fails to transfer, while a strong SWIR lens cannot compensate for a field of view that leaves the target represented by too few sensor pixels. The correct selection process therefore starts with sensor dimensions and pixel pitch, then connects them with lens image format, focal length and object-side resolution.

Start With Active Sensor Dimensions, Not the Sensor Format Name Alone

A camera specification may describe the detector using a format designation such as 2/3-inch, but the OEM should still obtain the actual active sensor width, height and diagonal whenever possible. Format names are useful compatibility references, yet real optical calculations require physical dimensions because field of view and image-circle coverage depend on millimetres rather than a format label.

The Kyptec Automation® SWIR range is currently specified for a 2/3-inch sensor format. When evaluating an InGaAs camera, the buyer should compare its active diagonal with the image format supported by the lens and verify that the complete required sensor area is covered. If the sensor extends beyond the useful image circle, corner darkening, missing image information or unacceptable edge performance can result. If the sensor is substantially smaller, coverage may be safe but the achievable field of view will differ from calculations made for the larger format.

Image Circle Determines Whether the Complete InGaAs Sensor Can Be Used

A lens projects a circular image onto a rectangular sensor. The sensor diagonal must fit within the lens's usable image circle if the entire active area is expected to receive valid image information. Sensor diagonal can be calculated from:

Sensor Diagonal = √(Sensor Width² + Sensor Height²)

If an InGaAs sensor measures 8.8 mm × 6.6 mm, for example, its diagonal would be approximately:

√(8.8² + 6.6²) ≈ 11.0 mm

The actual camera dimensions should always be used in the final calculation. This simple check is important because physical mount compatibility does not guarantee optical sensor coverage. A C-Mount lens can attach correctly while still being unsuitable for a sensor that requires a larger image field.

Pixel Pitch Determines How Finely the Sensor Samples the Optical Image

Pixel pitch is the physical center-to-center spacing between adjacent detector pixels, usually expressed in micrometres. Smaller pixels sample the optical image more finely, while larger pixels provide coarser spatial sampling. This does not mean that smaller pixels automatically produce better industrial inspection. The lens must transfer enough contrast at the corresponding spatial frequency, and enough SWIR signal must reach each pixel to support the required exposure and signal-to-noise ratio.

A useful theoretical relationship is the sensor Nyquist frequency:

Nyquist Frequency = 1 / (2 × Pixel Pitch)

where pixel pitch is expressed in millimetres.

For an illustrative 10 µm pixel:

10 µm = 0.010 mm

Therefore:

Nyquist Frequency = 1 / (2 × 0.010) = 50 cycles/mm

This means the sensor can theoretically sample spatial information up to approximately 50 cycles/mm before sampling limitations become dominant. It does not mean the complete camera system automatically resolves 50 cycles/mm with useful contrast. The lens's modulation transfer, focus, aperture, wavelength, motion and signal quality all determine how much practical information survives.

Why Pixel Pitch Matters More Than Megapixel Count Alone

Two cameras can both be described as 2 MP while using different sensor dimensions and pixel pitches. If one packs the same number of pixels into a smaller active area, the pixels will generally be smaller. That changes the spatial frequency the lens must support and also changes the FOV produced by a given focal length.

For this reason, choosing a “2 MP SWIR lens” for a “2 MP camera” is not a complete compatibility method. The megapixel designation is useful as an optical-resolution class, but a serious machine-vision design should additionally compare active sensor size, pixel pitch, required MTF, wavelength, object-side sampling and minimum defect size.

The current Kyptec Automation® portfolio is identified as a 2 MegaPixel SWIR lens range, including the Kyptec Automation® KL-1410 12.5 MM SWIR Camera Lens, which is specified for 900–1700 nm, 2 MP, 2/3-inch format, F1.4 and C-Mount. The buyer should interpret that specification together with the actual InGaAs sensor rather than as an independent guarantee of the smallest detectable defect.

Lens Resolution Must Preserve Contrast, Not Merely Separate Lines

Resolution in industrial imaging is fundamentally about whether useful contrast remains at the spatial scale required by the application. A lens may technically distinguish fine structures but transfer them with such weak contrast that an automated classifier cannot use the result reliably. This is why MTF—modulation transfer function—is more meaningful than a simple statement that a lens can “resolve” a particular feature. MTF describes how effectively contrast is preserved as object detail becomes finer.

Suppose a dark defect against a bright background begins with 80% contrast at the object. After illumination, material interaction, lens transfer, sensor sampling and processing, the recorded defect may retain only a fraction of that contrast. If the resulting difference approaches the level of production noise, the defect may become unreliable even though it occupies several pixels. Matching an InGaAs camera therefore requires enough optical contrast at the spatial frequency corresponding to the inspection target, not simply compatible megapixel labels.

Convert Pixel Pitch Into Object-Side Resolution

Sensor pixel pitch alone does not tell an OEM how small a defect can be inspected. The lens magnification or field of view converts sensor pixels into physical dimensions on the product. For many machine-vision applications, the simplest practical calculation is:

Object-Side Pixel Size = FOV / Number of Pixels Across That Axis

If an InGaAs camera provides 1600 horizontal pixels and the SWIR lens creates a 320 mm horizontal FOV:

320 ÷ 1600 = 0.20 mm/pixel

A 2 mm material feature would span approximately:

2 ÷ 0.20 = 10 pixels

If the same camera covers 800 mm:

800 ÷ 1600 = 0.50 mm/pixel

The same feature now spans only four pixels. Nothing about the sensor megapixel count has changed; the reduction comes entirely from the optical geometry. This is why focal length selection is inseparable from resolution planning.

Define the Minimum Defect Before Choosing the SWIR Lens

An OEM should never begin with “Which focal length is best for my InGaAs camera?” without first defining what the system must detect. The correct starting point is the smallest material region, defect, contaminant, crack, coating irregularity or other feature that must produce a reliable production decision. Once that target size is defined, the engineer can establish how many pixels should represent it and calculate the maximum permissible FOV.

For classification-only applications involving large homogeneous objects, relatively few spatial details may be required. For contamination detection or fine hidden-defect inspection, the target may need substantially stronger pixel representation. There is no universal number of pixels that guarantees detection because optical contrast, blur, noise and classification method also matter. The key is to create enough spatial margin that the smallest valid target is not operating at the sampling limit.

The 8.5 mm Option Is About Coverage, Not Simply “Wide Angle”

For broad inspection fields, the Kyptec Automation® KL-1408 8.5 MM SWIR Camera Lens can be evaluated when a large product area or conveyor width must fit into a limited working distance. The 8.5 mm focal length provides the widest geometry in the current Kyptec Automation® SWIR family, which can be valuable for material sorting, broad-area inspection and applications where each target remains physically large.

The trade-off is straightforward: a wider field allocates fewer pixels to each millimetre of product. Therefore, the Kyptec Automation® KL-1408 should be selected when its coverage advantage remains compatible with the minimum feature size rather than simply because it allows the greatest scene width.

Sensor Size Changes FOV Even When Focal Length Stays the Same

For the same focal length and working distance, a larger active sensor captures a wider field because more of the projected image is recorded. A smaller sensor captures a narrower central portion. This is why copying a focal length from one InGaAs camera to another can produce unexpected framing if their active sensor dimensions differ.

A simplified relationship for distant-object machine-vision planning is:

FOV ≈ Sensor Dimension × Working Distance / Focal Length

Although precise close-range calculations should use the complete lens geometry, the relationship illustrates the main dependence. Increase sensor width or working distance and the field becomes wider; increase focal length and the field becomes narrower. The actual product setup should always be validated physically before production release.

Pixel Count and Sensor Size Must Be Considered as a Pair

Suppose two InGaAs sensors both have 1600 horizontal pixels. Sensor A is 16 mm wide and Sensor B is 8 mm wide. Their approximate horizontal pixel pitches are:

Sensor A: 16 / 1600 = 0.010 mm = 10 µm

Sensor B: 8 / 1600 = 0.005 mm = 5 µm

Sensor B samples the image twice as finely at the sensor plane, but the lens must preserve useful contrast at correspondingly higher spatial frequencies to exploit that sampling. The smaller physical sensor will also produce a narrower FOV with the same lens and working distance. This demonstrates why pixel count by itself says surprisingly little about complete optical-system behaviour.

Smaller Pixels Place Greater Demands on Lens Performance

As pixel pitch decreases, the sensor can sample finer detail, but only if the optical image contains that detail with usable contrast. If the lens blur spot covers several pixels, reducing pixel pitch further may increase data volume without providing equivalent improvement in detectable object detail.

This is especially relevant in SWIR, where wavelength is longer than visible light and where material contrast, detector noise and optical transmission can matter as much as geometric sharpness. The correct lens should therefore be evaluated as part of the entire system rather than assuming that a smaller-pixel InGaAs camera always demands the highest nominal megapixel lens available.

Diffraction Also Connects Aperture, Wavelength and Pixel Pitch

Diffraction becomes increasingly relevant when the aperture is stopped down and wavelengths become longer. The approximate diameter of the central Airy disk can be represented as:

Airy Diameter ≈ 2.44 × λ × F-number

At a wavelength of 1.5 µm and F4, the approximate diameter is:

2.44 × 1.5 × 4 ≈ 14.6 µm

This does not directly predict complete industrial image quality, but it demonstrates why SWIR optical resolution cannot be discussed without wavelength and aperture. If a sensor uses very small pixels, several pixels may lie inside the diffraction spot when the lens is stopped down substantially. Conversely, operating wide open can improve diffraction-limited resolution and light collection but may reduce depth of field and expose other optical aberrations. The production aperture should therefore be selected from actual imaging performance, not from one theoretical parameter.

Why F1.4 Matters for InGaAs Machine Vision

The Kyptec Automation® SWIR range uses an F1.4 maximum aperture across the current portfolio. This is useful because many SWIR inspection systems operate under lower available photon levels than bright visible-light imaging, particularly when narrow spectral bands or short exposure times are required. Opening the aperture can increase collected signal and support shorter exposures on moving production lines.

However, a wider aperture is not automatically the optimum setting. Product-height variation may require more depth of field, and certain applications may benefit from stopping down to improve focus tolerance. The buyer should therefore regard F1.4 as available optical throughput and then qualify the final aperture according to exposure, focus margin and object geometry.

Sensor Pixel Pitch Must Be Connected With Signal-to-Noise Ratio

Smaller pixels collect light over a smaller physical area, all else being equal. In a demanding SWIR application, the finest spatial sampling is not useful if each pixel receives too little signal to distinguish the target from noise. Increasing exposure can improve signal but may introduce motion blur; stronger illumination can help but may increase system cost or thermal constraints; opening the aperture improves light collection but changes depth-of-field behaviour.

The correct InGaAs-camera and SWIR-lens pairing therefore balances spatial sampling with usable signal, rather than treating minimum pixel pitch as an independent goal.

Working Distance Determines More Than Mechanical Clearance

Working distance changes the field of view and therefore the number of pixels covering the product. It can also affect illumination geometry, mechanical access and focus tolerance. A machine that places the camera unnecessarily far away may require a longer focal length to recover the desired FOV, while a very short working distance can create lighting and mechanical interference.

The Kyptec Automation® SWIR family provides several focal lengths so the machine builder can solve these geometric constraints without forcing one optical configuration onto every application. This is an important practical advantage for OEMs designing different machine sizes around the same SWIR sensor platform.

Kyptec Automation® KL-1412 for Controlled 2/3-Inch InGaAs Inspection

The Kyptec Automation® KL-1412 25 MM SWIR Camera Lens can be a useful choice where a 2/3-inch InGaAs camera is inspecting one product or a controlled region rather than a broad conveyor. Its tighter field can place more of the available 2 MP sampling on the region that actually determines the production decision. This makes 25 mm particularly worth evaluating when small material features need stronger spatial representation but the application does not require the narrower geometry of the longest focal lengths.

Sensor Cropping Should Not Be Used to Hide Poor Lens Selection

It is possible to crop an image electronically to remove dark or low-quality corners, but doing so reduces the effective active sensor area and changes the usable FOV. If significant cropping is required merely to make the optical system acceptable, the lens-sensor match should be reconsidered.

A better design uses the required active sensor area intentionally. If the application needs only a small region, choosing a focal length that fills more of the sensor with useful product information is often preferable to capturing an unnecessarily wide field and discarding most of it afterward.

Edge Performance Matters When the Full Sensor Is Used

An InGaAs system can provide excellent center sharpness while producing weaker inspection performance near the image edges because of illumination falloff, residual aberration, focus variation or geometric effects. If every part of the sensor contributes to the production decision, the smallest defect should be validated across the full usable field.

This is particularly important when the sensor format is close to the supported lens format. Qualification should use actual defects or representative targets at the center, edges and corners instead of assuming center performance describes the complete image.

C-Mount Compatibility Is Necessary but Not Sufficient

The current Kyptec Automation® SWIR portfolio uses C-Mount. This provides a familiar mechanical interface for many industrial camera configurations, but mount compatibility addresses only one part of integration. The OEM must still verify flange geometry, sensor location, focus range, image-circle coverage, available mechanical clearance and the camera's specific optical interface.

A lens that screws onto the camera is not necessarily a correctly matched imaging system. Mechanical compatibility should be checked first, but optical compatibility must still be proven through calculated and physical testing.

Broadband 900–1700 nm Operation Requires Wavelength-Aware Qualification

If an InGaAs inspection system uses several wavelengths, lens performance should be checked at the spectral regions that actually drive the production decision. Focus, material contrast, signal level and sensor response can all vary with wavelength. A setup optimized under one narrow band should therefore not automatically be assumed equivalent at another.

The dedicated Kyptec Automation® SWIR Camera Lens collection is specified for 900–1700 nm operation, which gives OEMs an appropriate foundation for broadband SWIR integration. Final application performance should still be verified at the chosen illumination wavelengths because complete system behaviour includes the camera, lens, illumination and inspected material.

Do Not Choose the 50 mm Lens Simply Because It Gives “More Detail”

A longer focal length narrows the FOV at a given sensor size and working distance, which can increase pixels per millimetre on the product. It does not independently increase the intrinsic sensor resolution or make the material more spectrally distinguishable. The Kyptec Automation® KL-1416 50 MM SWIR Camera Lens is therefore most appropriate when a relatively small field or greater stand-off matches the machine geometry. The live product page identifies it as a 50 mm, 900–1700 nm, 2 MP, 2/3-inch SWIR lens. Its advantage is tighter framing, not an automatic guarantee of better inspection.

Match the Lens to the Smallest Commercially Relevant Feature

The final lens selection should always return to the production requirement. Suppose the buyer needs to classify entire 50 mm polymer pieces. A broad field may still provide abundant pixels per object. If the real requirement is detecting a 0.5 mm contaminant, the optical design becomes much more demanding. The same 2 MP InGaAs camera can therefore require completely different focal lengths in those two machines.

This is why a procurement specification should state minimum feature size alongside camera resolution. “2 MP InGaAs camera with 2 MP SWIR lens” describes the components; “detect a 1 mm target over a 250 mm FOV at 300 mm working distance” describes the machine-vision requirement.

Why Kyptec Automation® Is a Strong Platform for InGaAs Camera Integration

Kyptec Automation® provides a focused SWIR Camera Lens portfolio with five focal lengths from 8.5 mm to 50 mm rather than a single fixed geometry. The live collection currently lists exactly five products, while representative product pages specify 900–1700 nm wavelength coverage, 2 MP resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount. For OEMs, this means a common InGaAs camera architecture can potentially be paired with different Kyptec Automation® focal lengths depending on whether the machine requires broad conveyor coverage, balanced field and sampling, tighter individual-product inspection or greater working-distance flexibility.

The practical strength of the portfolio is therefore not simply the number of focal lengths. It is the ability to approach lens selection systematically: verify the sensor format, understand pixel pitch, define the smallest target, calculate the required object-side sampling, determine the necessary FOV and working distance, then choose the focal length that satisfies those requirements with useful production margin.

Frequently Asked Questions About SWIR Camera Lenses for InGaAs Cameras

1. What camera specifications should I check before buying a SWIR lens for an InGaAs sensor?

Start with the active sensor width, height and diagonal, pixel count, pixel pitch, lens mount and spectral response range. Then define the required field of view, working distance and smallest inspection feature. These parameters determine whether the lens will cover the sensor and whether the complete optical system can provide enough spatial sampling for the application.

2. Is sensor diagonal or sensor width more important when checking SWIR lens compatibility?

Both serve different purposes. Sensor diagonal is important for checking whether the lens image field can cover the complete sensor, while sensor width and height are needed to calculate horizontal and vertical FOV. A robust design uses all three physical dimensions rather than relying only on the nominal format name.

3. Why can two 2 MP InGaAs cameras require different SWIR lens performance?

Because megapixel count does not define sensor size or pixel pitch. Two 2 MP sensors can distribute their pixels across different physical areas, producing different pixel sizes, Nyquist frequencies and FOVs with the same lens. Lens matching should therefore use physical sensor specifications in addition to pixel count.

4. What does Nyquist frequency tell me when choosing a SWIR lens?

Nyquist frequency describes the highest spatial sampling frequency supported theoretically by the sensor based on its pixel pitch. It helps indicate how demanding the sensor is on the lens. Practical image quality also depends on lens MTF, wavelength, aperture, focus, signal-to-noise ratio and production motion, so Nyquist should be treated as one part of the compatibility analysis.

5. Can a lens rated for fewer megapixels than the camera waste sensor resolution?

Potentially, because the camera may sample more finely than the optical image contains useful contrast. The exact effect depends on pixel pitch and lens MTF rather than megapixel labels alone. A higher camera pixel count provides limited benefit if optical blur or weak contrast dominates before the sensor sampling limit is reached.

6. Does a 2 MP SWIR lens guarantee detection of a specific defect size?

No. The 2 MP designation describes the lens resolution class, not an application-specific detection guarantee. Minimum defect detection also depends on FOV, number of camera pixels, contrast, focus, motion blur, wavelength, illumination and classification margin. Defect size should always be converted into expected pixels at the final machine geometry.

7. How do I calculate millimetres per pixel for an InGaAs inspection system?

Divide the physical field of view along one axis by the number of active pixels along the same axis. For a 240 mm FOV and 1200 pixels, the sampling is 0.20 mm/pixel. A 2 mm object would then occupy approximately ten pixels before accounting for blur and other optical effects.

8. Can an InGaAs sensor be smaller than the lens's specified image format?

Yes. Using a smaller sensor normally means it captures a smaller central region of the lens image, so sensor coverage itself is generally easier. However, the FOV will become narrower than on the larger format, and the complete lens-camera combination still needs to meet spatial-resolution and working-distance requirements.

9. Why does my SWIR camera have enough pixels but still fail to resolve the target?

Possible causes include an excessively wide FOV, insufficient lens contrast at the required spatial frequency, poor focus, motion blur, diffraction, low signal-to-noise ratio or weak material contrast. Pixel count represents sampling capacity, but useful target detection requires the complete optical chain to preserve the target information.

10. Should pixel pitch influence aperture selection on a SWIR system?

Yes, particularly when fine pixels are combined with longer SWIR wavelengths. Stopping down increases depth of field but also increases diffraction, which can reduce contrast at fine spatial frequencies. The final aperture should therefore be tested against the actual pixel pitch and inspection target rather than selected from depth of field alone.

11. How does lens focal length affect the use of a 2 MP InGaAs sensor?

Focal length changes the physical FOV recorded by the sensor at a given working distance. Shorter focal lengths generally provide wider fields and fewer pixels per millimetre, while longer focal lengths provide narrower fields and greater spatial sampling of the target. The camera still has the same pixel count; what changes is how those pixels are distributed over the product.

12. When is the Kyptec Automation® KL-1408 appropriate for an InGaAs camera?

The Kyptec Automation® KL-1408 8.5 MM SWIR Camera Lens is particularly relevant where broad field coverage is required and the targets are large enough to remain well sampled. The OEM should calculate pixels per minimum feature before selecting the wide field as the final configuration.

13. When should an OEM consider the Kyptec Automation® KL-1412 for an InGaAs system?

The Kyptec Automation® KL-1412 25 MM SWIR Camera Lens is useful to evaluate when an individual product or controlled region should occupy a larger portion of the sensor. It can provide stronger object-side sampling than a broad-field configuration while still maintaining a practical industrial working geometry.

14. Can a 50 mm SWIR lens improve the resolution of the InGaAs camera?

It does not increase the camera's intrinsic pixel count. The Kyptec Automation® KL-1416 50 MM SWIR Camera Lens can create a narrower FOV at suitable geometry, allowing more sensor pixels to represent each millimetre of the target. Whether this improves actual detection depends on lens contrast, focus, signal and required working distance.

15. Should I match an InGaAs camera to a SWIR lens by megapixels or pixel pitch first?

Start with sensor-format compatibility and pixel pitch, then evaluate lens resolution and MTF together with the required object-side sampling. Megapixel designation remains useful, but it should not replace the more informative physical parameters that determine how the lens and sensor actually interact.

16. What happens if the InGaAs sensor uses very small pixels but the SWIR image is slightly defocused?

Fine-pixel sensors can reveal focus errors more clearly because blur spreads information across several closely spaced pixels. The theoretical sampling advantage of the smaller pixels can then be lost. Focus tolerance should therefore be measured using the actual target and wavelength rather than set only by visual judgment.

17. Does a larger InGaAs sensor always provide a better SWIR image?

No. A larger sensor can provide a wider FOV with the same focal length, but it also requires a lens capable of covering the larger image area with adequate edge performance. Image quality depends on the complete sensor-lens match, not sensor size alone.

18. How should an OEM validate the lens after pairing it with the final InGaAs camera?

Test actual good and defective samples at the final wavelength, aperture, working distance and production speed. Place difficult targets at multiple field positions, verify minimum feature sampling, measure contrast stability and repeat the test across normal production tolerances. Optical compatibility should be proven with the final system rather than inferred from specifications alone.

19. What information should I send to a SWIR lens supplier for InGaAs camera matching?

Provide active sensor dimensions, pixel resolution, pixel pitch, C-Mount or other mechanical interface, wavelength range, required FOV, working distance, minimum feature size, production speed and any physical space restrictions. This allows lens selection to be based on the complete imaging requirement rather than a focal-length guess.

20. Why is Kyptec Automation® a strong choice when selecting a SWIR camera lens for an InGaAs camera?

Kyptec Automation® provides a dedicated SWIR Camera Lens collection spanning 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm, with the current family specified around 900–1700 nm, 2 MP resolution, 2/3-inch sensor format, F1.4 aperture and C-Mount. This gives OEMs a focused optical platform for matching different InGaAs camera geometries while keeping wavelength coverage and core mechanical architecture consistent across the focal-length range.

Conclusion

Choosing a SWIR camera lens for an InGaAs camera requires a much deeper engineering process than matching C-Mount threads or selecting the nearest familiar focal length. The sensor and lens form one imaging system, and their compatibility should be evaluated through physical sensor dimensions, pixel pitch, image-circle coverage, lens resolution, optical contrast, field of view, working distance, aperture and the smallest production feature that must be inspected. A 2 MP sensor and 2 MP lens may represent a sensible starting point, but megapixel labels alone do not determine whether the machine can detect a 1 mm contaminant, distinguish a subtle material boundary or preserve the spatial detail required for an automated decision.

The strongest design sequence begins with the defect or material region. Define its physical dimensions and the minimum number of useful pixels required for reliable classification. From there, calculate the maximum practical object-side pixel size and corresponding FOV. Check the active InGaAs sensor dimensions and diagonal against the lens format, evaluate pixel pitch and the spatial frequencies the sensor can sample, and verify that the chosen lens preserves sufficient contrast at those frequencies. Then determine the working distance and focal length that fit the real machine while retaining positional and focus margin. Aperture should be optimized only after signal, diffraction, depth of field and production exposure requirements are understood.

The Kyptec Automation® SWIR Camera Lens collection provides a particularly useful platform for this engineering approach because the OEM can choose among 8.5 mm, 12.5 mm, 25 mm, 35 mm and 50 mm focal lengths while staying within a dedicated 900–1700 nm, 2 MP, 2/3-inch, F1.4 and C-Mount family. Shorter focal lengths can support broader fields, intermediate choices can balance coverage and spatial sampling, and longer focal lengths can allocate more sensor pixels to a smaller inspection area when working distance permits. This gives machine builders meaningful flexibility to match the optical field to the actual InGaAs sensor and application instead of forcing a single focal length onto every SWIR system.

For an OEM or industrial buyer, the central rule is therefore straightforward: do not buy a SWIR lens because its megapixel rating appears to match the camera; buy it because the complete lens–sensor combination delivers the required image circle, spectral transmission, spatial contrast and pixels on the smallest commercially important feature at the real production working distance. When sensor size, pixel pitch, 2 MP optical performance, focal length and SWIR wavelength behaviour are engineered together, Kyptec Automation® SWIR Camera Lenses provide a strong foundation for building reliable InGaAs machine-vision systems across the 900–1700 nm range.