LightingGuide

Infrared flood illumination for stereo depth imaging

Choose infrared flood illumination for stereo depth: compare coverage, exposure, ambient rejection and texture, then verify timing and safety.

LumaSpec Vision10 min read Share on LinkedIn

On this page 11 sections
  1. 01Decide whether you need brightness or texture
  2. 02Specify wavelength and optical requirements first
  3. 03Calculate coverage at the near and far limits
  4. 04Check whether the texture is large enough to resolve
  5. 05Position the source without obstructing either view
  6. 06Test exposure and ambient rejection quantitatively
  7. 07Verify range and synchronization independently
  8. 08Treat compact catalog parts as documentation starting points
  9. 09Run an acceptance matrix before ordering
  10. 10Finish with safety and purchase checks
  11. 11Questions

Uniform infrared lighting can brighten a stereo image without improving the depth map. The difference is usually whether the scene contains usable texture, whether both imagers see it clearly, and whether the added light reaches the target during exposure.

For a compact inspection station, choose the illumination geometry before choosing the emitter technology. This guide explains how to specify coverage, position a small source, compare exposure settings, and verify useful depth rather than image brightness alone. It also separates catalog information from the additional documentation needed to select a compatible infrared flood source.

Decide whether you need brightness or texture

Stereo depth estimates distance by matching features between two camera views. A uniform flood raises scene brightness, but it does not deliberately add spatial features. A smooth, uniformly reflective surface can therefore remain difficult to match after illumination increases.

A pattern projector serves a different purpose: it adds visible structure in the cameras' operating band. Replacing that pattern with a uniform field changes the matching conditions, even if the total emitted optical power remains unchanged.

Scene or symptom What a uniform flood may improve What it cannot reliably solve
Textured surface with dark images Signal level and potentially shorter exposure Incorrect calibration or timing
Smooth, featureless surface Image brightness Lack of distinguishable stereo features
Dark material with weak texture Visibility of existing texture Very low reflectance at the chosen wavelength
Reflective curved surface Some underexposed regions View-dependent reflections and saturated highlights
Strong ambient illumination Active signal within the exposure Unlimited rejection of ambient light

Start with the stereo lighting checks for dark surfaces when material reflectance is the main limitation. If only particular regions lack depth, the depth-hole diagnosis workflow helps distinguish lighting problems from occlusion and geometry.

Specify wavelength and optical requirements first

Confirm the supported spectral range of both stereo imagers before selecting a light. Candidate wavelengths such as 850 nm and 940 nm are not interchangeable: sensor sensitivity, installed filters, target reflectance, and available source output can differ.

Write a short requirement sheet containing:

  • The wavelength and spectral bandwidth accepted by both imagers.
  • The minimum and maximum target distance.
  • The required scene width and height at those distances.
  • The longest acceptable exposure and the intended acquisition rate.
  • The darkest relevant target material and its surface finish.
  • The permitted source envelope and mounting position.
  • The required operating mode, trigger interface, and safety documentation.

Specify radiant output or target irradiance where documentation provides it. A listed power value alone does not establish how much useful infrared light reaches the scene.

LED and vertical-cavity surface-emitting laser, or VCSEL, sources can have different optical and safety requirements. Neither label guarantees uniform coverage, sufficient range, or compatibility with a camera. Select the complete illumination assembly, not just its emitter technology.

Calculate coverage at the near and far limits

The flood must cover the common viewing region of both stereo imagers. Check that region at the nearest target plane, where a fixed beam has its smallest footprint, as well as at the farthest plane.

The field-of-view diagram accompanies practical positioning guidance for depth cameras. Keep camera viewing geometry separate from the illuminator's beam geometry.

How sensor width, working distance and field of view set the focal length Rays from the edges of the sensor pass through the lens and spread to the edges of the field of view. The ratio of sensor width to field of view is the magnification m.CameraField of view (FOV)Working distance (WD)≈ fsensor width slensm = s / FOVf = WD × m / (1 + m)
How sensor width, working distance and field of view set the focal length. Rays from the edges of the sensor pass through the lens and spread to the edges of the field of view. The ratio of sensor width to field of view is the magnification m.

For a centered source with an approximately conical beam, estimate the illuminated diameter using:

Beam diameter = 2 × distance × tan(full beam angle / 2)

Suppose a hypothetical source has a full beam angle of 60 degrees. At a distance of 0.8 m:

Beam diameter = 2 × 0.8 × tan(30 degrees) ≈ 0.92 m

At 1.6 m, the estimated diameter becomes 1.85 m. These are geometric estimates, not guaranteed uniform regions. Beam-angle definitions vary, and substantial brightness variation may exist inside the stated footprint.

For a rectangular scene, check the corners rather than only its width. A scene measuring 1.2 m by 0.8 m has a diagonal of:

Scene diagonal = sqrt(1.2² + 0.8²) ≈ 1.44 m

A centered circular footprint must cover that diagonal before alignment margin is added. Measure the usable illuminated area on a representative target rather than accepting beam diameter alone.

Check whether the texture is large enough to resolve

Brighter images help only if relevant features survive sampling and focus. As a first estimate, divide scene width by horizontal image pixels.

For a hypothetical image covering 1.2 m across 1280 pixels:

Object sampling = 1200 mm / 1280 pixels ≈ 0.94 mm per pixel Width of a three-pixel feature = 3 × 0.94 mm ≈ 2.8 mm

This does not establish a universal minimum feature size for stereo matching. It shows why fine markings may disappear when the field becomes wider. Blur, low contrast, repetitive patterns, and the matching algorithm also matter.

Camera resolution for the smallest feature
Units
Pixels needed
–
Sensor size
– MP
Smallest sensor that fits
–
Pixel on the part
– mm
Pixels across the feature, that sensor
–

Pixels needed = field / feature × pixels per feature on each axis (field incl. margin). The lens must resolve the pixels too: Nyquist = 1000 / (2 × pixel µm) lp/mm.

Find cameras with enough pixels Open the full calculator

Use the calculator to check image sampling, then inspect actual left and right frames. Look for recognizable, unsaturated texture in both views. A bright image with barely distinguishable surface detail is not a successful lighting result.

Position the source without obstructing either view

A small source close to the stereo assembly can reduce some illumination shadows. It does not eliminate occlusions between the two camera viewpoints.

Place the light where its housing, cable, and mounting hardware remain outside both viewing cones. Check this at the nearest working distance and throughout the allowed part positions. Do not place an external source over an integrated projector window unless the camera manufacturer explicitly permits that arrangement.

Start near the center of the stereo assembly, then compare an offset position. A centered position may reduce asymmetric shadows, while an offset position may move troublesome reflections away from the imagers. The better choice depends on the inspected surface.

Check these conditions separately:

  • A tall feature beside a recessed region.
  • A reflective face tilted toward the source.
  • A dark surface at the far distance limit.
  • A target near each corner of the common viewing region.

If both images lose contrast after the source moves closer to the camera, investigate direct light entering the lenses. The stray-light troubleshooting guide explains why more illumination can sometimes produce less useful contrast.

Test exposure and ambient rejection quantitatively

Disable automatic exposure and gain where supported during comparison tests. Otherwise, the camera may change settings between trials and hide the actual contribution of the external light.

Capture ambient-only and light-on images with identical settings and a stationary scene. Use raw or linear image data when available. Measure the same unsaturated target patch in each image.

For example, suppose an ambient-only patch averages 40 digital counts and the light-on patch averages 120 counts:

Estimated active contribution = 120 - 40 = 80 counts Active-to-ambient ratio = 80 / 40 = 2

This is an image-level diagnostic, not a calibrated irradiance measurement or a complete signal-to-noise calculation. It becomes misleading when images are saturated, processed nonlinearly, or captured with changing gain.

A matched bandpass filter can reduce out-of-band ambient light. It also reduces some useful signal, and sunlight contains infrared energy within common passbands. Confirm filter compatibility, transmission, and angle sensitivity for both imagers. Follow the camera manufacturer's guidance before changing optical components or calibration.

The ambient-rejection measurement guide develops this comparison further.

Verify range and synchronization independently

For a source behaving approximately as a point source in its far field, target irradiance falls roughly with the inverse square of distance. Doubling distance therefore reduces irradiance to about one quarter.

Under otherwise unchanged conditions, recovering the same collected signal might require approximately four times the exposure. This is only an estimate: real beam optics, surface angle, reflectance, and camera processing can change the relationship.

Longer exposure also increases motion blur. If an object moves at 0.5 m/s during a 2 ms exposure:

Object displacement = 0.5 m/s × 0.002 s = 0.001 m = 1 mm

Compare that displacement with the object sampling and the texture size needed for matching. Increased brightness is valuable when it permits shorter exposure without losing useful signal.

For pulsed operation, use only documented camera outputs and compatible controller inputs. A timing output is not a power connection for an illuminator. Verify signal levels, polarity, pulse duration, and the exposure interval of both imagers.

Do not assume a frame synchronization output represents the complete exposure window. Rolling-shutter operation may require illumination across a longer interval than an individual row exposure.

Use the stereo synchronization guide for the timing investigation, and the light-controller duty-cycle guide before setting pulse limits.

Treat compact catalog parts as documentation starting points

The spot lights category and lighting finder provide starting points for comparing physical formats. The supplied catalog information does not establish either example below as a compatible infrared flood source.

The LS-SPT-0133 spot light lists power 4 W. Its supplied entry does not establish wavelength, beam uniformity, or stereo compatibility.

Spot lightsThird generation spot light, power 4 W (LS-SPT-0133)LS-SPT-0133
Size
Under 50 mm
Voltage
24 V
$259.00View partBuild a system with it

The LS-SPT-0175 spot light lists power 5 W. That value alone does not establish delivered infrared irradiance.

Spot lights6 mm High luminance spot light, power 5 W (LS-SPT-0175)LS-SPT-0175
Size
Under 50 mm
$229.00View partBuild a system with it

Do not rank these parts by the listed power and assume the higher value will solve a range problem. Request wavelength, optical distribution, continuous and pulsed operating limits, and safety information before selecting an illumination candidate.

Review lighting accessories through the accessories finder after the light and controller interfaces are defined.

The LS-LAC-0024 light cable is an accessory to review against the selected interfaces. Its inclusion here does not establish compatibility with either light.

Lighting accessoriesLight Cable, 19M-2E to 19M-2A, 5 m (LS-LAC-0024)LS-LAC-0024
Size
Under 50 mm
$49.00View partBuild a system with it

The LS-LAC-0060 extension cable also requires a connector, pinout, and electrical-rating check before selection.

Lighting accessories5 m light extension cable (LS-LAC-0060)LS-LAC-0060
Size
Under 50 mm
$55.00View partBuild a system with it

If the task requires a different depth-sensing method, compare the 3D sensors category using the camera finder. Changing the measurement approach may be more effective than increasing flood intensity on an untextured target.

Run an acceptance matrix before ordering

Test useful depth under the worst relevant combinations, not only at the center of a convenient target.

Variable Suggested comparison What to record
Distance Near, nominal, and far Valid-depth fraction and depth error
Surface Dark, textured, smooth, and reflective Missing regions and false matches
Ambient light Lowest and highest expected levels Active contribution and saturation
Motion Stationary and maximum production speed Blur and depth instability
Source position Centered and practical offset Shadows, glare, and coverage
Exposure Shortest useful through longest permitted Noise, clipping, and valid depth

Define valid-depth fraction as valid pixels divided by all pixels inside the inspection region. For example, 18,000 valid pixels in a 20,000-pixel region gives 90%. Set the acceptance threshold from the inspection requirement; 90% is not automatically sufficient.

Also measure error against a known distance or reference geometry. A dense depth map can still be wrong. With a stationary target, record variation across a sequence as well as average error. Keep that test separate from motion testing.

For a longer run, follow the lighting stability test workflow. Repeat the critical measurements after the system reaches its normal operating condition.

Finish with safety and purchase checks

Invisible infrared radiation is not safe merely because it appears dim or cannot be seen. LED and laser-based assemblies require the appropriate documented assessment. Applicable references may include IEC 62471 for lamp systems and IEC 60825-1 for laser products; applicability depends on the complete assembly.

Use the supplier's documented viewing distances, operating limits, and protective measures. Do not remove diffusers, alter beam-forming optics, exceed current limits, or defeat protective functions to obtain more range. An average-power calculation alone is not a pulsed-operation safety assessment.

Before committing to parts, confirm:

  • Both imagers support the selected wavelength and filter arrangement.
  • The useful beam covers the full common scene, including its corners.
  • Natural texture remains visible at the required exposure.
  • The far-distance test passes under the highest expected ambient level.
  • Timing interfaces and pulse limits are documented and compatible.
  • Cables match the actual connectors, pinouts, and electrical requirements.
  • Optical safety documentation covers the intended operating configuration.

Start with the lighting configurator once those requirements are defined. If wavelength, coverage, or interface data remains missing, send the requirement sheet through application support before treating a compact spot light as an infrared flood solution.

Questions

Will uniform infrared flood illumination improve every stereo depth map?

It can improve depth when insufficient illumination hides existing surface texture. It does not deliberately create matching features, so smooth or uniformly reflective surfaces may remain difficult. Compare valid depth and measurement error rather than judging success from brighter images alone.

Should I choose 850 nm or 940 nm illumination?

Choose from the camera's documented spectral response, installed filters, target reflectance, and available illumination output. Neither wavelength is universally preferable. Verify both stereo imagers and test representative materials with the intended filter arrangement before committing to a source.

Is the smallest infrared emitter usually the best choice?

A compact source is useful only if it also provides adequate coverage, useful target irradiance, compatible timing, and documented safe operation. Emitter dimensions alone do not describe the complete illuminator. Beam-forming optics, mounting hardware, and required viewing clearances also affect the installed footprint.

Can a camera synchronization output drive an external light directly?

Use only a documented timing output connected to a compatible controller input. Do not use a camera timing signal as the illuminator's power supply. Check voltage levels, polarity, pulse limits, and how the signal relates to both imagers' actual exposure intervals.

Are the example spot lights confirmed infrared flood sources?

Not from the supplied catalog information. The example entries do not establish infrared wavelength, flood uniformity, or compatibility with a particular stereo camera. Request the missing optical, electrical, timing, and safety documentation before selecting either example for this application.