Strobed lighting ambient rejection: measure and improve it

A bright strobe does not automatically reject ambient light. The camera still collects background illumination for its entire exposure, including time before and after the pulse. For inspection stations affected by changing room lighting, the useful adjustment is often a shorter exposure rather than a higher drive setting.

This procedure separates ambient and strobe contributions, calculates their ratio, and checks whether exposure can be shortened without losing useful signal. It also identifies the light-and-controller information you need before considering higher pulse intensity.

1. Define what ambient rejection means

Use the ratio of useful strobe signal to ambient signal as a practical measurement. Both signals must come from the same image region, with unchanged camera settings and part position.

These equations assume linear image values, stable ambient illumination between captures, and no clipping. The ratio describes illumination dominance, not the signal-to-noise ratio or inspection accuracy.

A ratio of 4 means the measured strobe contribution is four times the ambient contribution. The ambient fraction of the combined signal is then 1 / (4 + 1) = 20%. Decide what ratio is acceptable by testing your actual measurement, classification, or reading task.

Choose at least two regions: the feature being inspected and a representative background. A strong average ratio can conceal a reflection that saturates a small but important area.

2. Prepare a repeatable image measurement

1. Fix exposure, gain, aperture, focus, part position, and light geometry. Disable automatic brightness adjustments during the comparison.
2. Use linear image data where available. Disable gamma, tone mapping, and adaptive contrast processing that would invalidate subtraction.
3. Capture a dark reference with the lens covered, using the same exposure and gain. Repeat this reference after changing exposure.
4. Capture ambient-only images with the inspection light disabled through its normal control interface.
5. Capture strobe-lit images without moving the target or changing the camera settings.
6. Inspect individual pixels for clipping before calculating regional means.

For an initial stationary-target test, collect 20 paired captures at each setting. Record the median signal and the range across captures rather than trusting one convenient frame. This is a starting sample size, not production acceptance evidence.

Artificial lighting can fluctuate between paired captures. Interleave ambient-only and strobe-lit frames where practical, and repeat the test across different trigger phases. If brightness varies periodically, follow the diagnostic approach in camera flicker under artificial lighting.

If brightness changes even with fixed illumination, first check the acquisition settings using industrial camera image quality settings.

3. Establish exposure and pulse timing separately

Treat camera exposure and light pulse duration as different controls. The exposure determines how long ambient light accumulates. The pulse duration determines how long the inspection light contributes, provided the complete optical pulse falls inside the exposure.

For steady ambient light and a linear sensor:

Here, T is camera exposure, t_p is optical pulse duration, and E_p is pulse irradiance at the inspected region. The constants include the optical path, surface response, and sensor response. These expressions are estimates to test, not substitutes for image measurements.

Shortening exposure reduces ambient signal while preserving strobe signal only when the useful pulse remains fully captured. Include measured light startup, decay, trigger latency variation, and camera timing uncertainty when setting margins.

For example, a measured optical pulse lasting 100 microseconds could fit within a 200 microsecond exposure with 50 microseconds on either side. Those margins are illustrative; they are acceptable only if the measured timing variation fits inside them.

A global shutter simplifies this overlap requirement. With a rolling shutter, a short pulse can illuminate only some rows unless the camera provides a suitable common exposure interval or documented synchronization mode.

Use strobe overdrive, duty cycle and timing for the broader timing checks. If the first image differs from later images, investigate exposure settling and light timing before changing pulse intensity.

4. Shorten exposure and calculate the improvement

Start by retaining the same pulse setting and reducing camera exposure. This isolates ambient rejection from any change in light output.

Consider a stationary test using a 10-bit linear image, whose maximum digital value is 1023. The following are hypothetical regional measurements, not product specifications.

The arithmetic is:

The total image became darker, but the useful strobe contribution stayed constant. That is the intended outcome. Raising gain to restore the original overall brightness would not improve the optical rejection ratio.

If the measured strobe contribution falls, check pulse overlap before increasing drive. A slow optical rise, excessive trigger variation, or a partially captured pulse can invalidate the expected improvement. The tests in diagnosing strobe pulse startup and current droop help distinguish electrical and optical limitations.

When ambient contribution approaches the uncertainty of the dark reference and image measurements, do not report a precise, extremely large ratio. Report the measurement limit and validate performance directly.

5. Change pulse intensity only when measurements justify it

If the shorter exposure cannot contain the existing pulse, you may need a shorter, brighter pulse. That requires documented compatibility between the light, controller, wiring, and operating sequence.

Do not assume that multiplying commanded current multiplies emitted light by the same amount. Actual output depends on the LED assembly, temperature, electrical limits, and pulse startup behavior. A current waveform confirms electrical delivery; an image or optical detector measures the light that matters.

Suppose measured pulse irradiance doubles while pulse duration halves. The approximate integrated strobe signal remains unchanged because 2 × 0.5 = 1. If camera exposure also halves, steady ambient contribution halves. The strobe-to-ambient ratio therefore doubles, rather than quadruples.

Check duty cycle using the actual pulse repetition frequency:

Low duty cycle alone does not authorize a particular peak current. Maximum pulse duration, peak current, repetition rate, burst behavior, and thermal conditions must all comply with the documented limits.

6. Select the controller and light as a documented pair

Use the light controllers category to identify the required output architecture, then compare candidates in the controller finder. Confirm voltage or current compatibility, channel loading, trigger interface, and timing behavior before ordering.

The LS-CTL-0070 is cataloged with a strobe feature and constant-voltage output. Treat it as a candidate for compatibility review, not as evidence of a particular overdrive capability.

The supplied catalog information does not establish minimum pulse width, trigger latency, optical rise time, or allowable peak pulse current for this candidate. Obtain the relevant documentation through the product support process before specifying those requirements.

Choose light geometry for feature contrast before chasing higher intensity. Bar lights let you test directional illumination, while ring lights provide illumination around the lens. Use the lighting finder to narrow the physical arrangement.

The LS-BAR-1047 is a standard bar-light candidate for directional-lighting trials.

The LS-RNG-0143 is a ring-light candidate with a diffuser for geometry trials around the viewing axis.

Neither candidate's supplied catalog entry establishes a permissible overdrive envelope. A shared nominal voltage does not, by itself, prove compatibility with a controller's pulsed output.

If the inspection feature lacks contrast even with stable exposure, revisit the lighting selection test plan. More light cannot reliably repair unsuitable illumination geometry.

7. Verify rejection under the conditions that matter

Repeat the paired-image measurements after the station reaches a stable operating temperature. Include the darkest and brightest expected ambient conditions, the full trigger-rate range, and startup after an idle period.

Keep a test record containing exposure, gain, pulse command, measured optical timing, trigger rate, ambient contribution, strobe contribution, clipped-pixel count, and the inspection decision. This makes it easier to distinguish an optical change from a camera-setting change.

Check the actual feature measurement as well as image brightness. A ratio that looks acceptable on a matte reference may fail on a reflective part because the two light sources illuminate different surface directions.

Ambient light can also produce a moving ghost around an otherwise sharp strobed feature. Evaluate production-speed images, not just stationary samples. If residual ambient affects edge location or classification, reduce it further or change the optical arrangement.

For repeatability testing beyond this initial comparison, use the procedure in machine vision lighting stability before deployment.

8. Know when shorter exposure is the wrong adjustment

A shorter exposure will not help if it removes useful pulse energy as quickly as ambient energy. This can happen when the light takes too long to reach its useful output or when the camera lacks a suitable exposure interval.

It also cannot separate an unwanted light pulse that is synchronized with your exposure. In that case, investigate source scheduling, optical shielding, or spectral separation instead. A lens filter may help when the unwanted illumination has a sufficiently different spectrum, but verify transmission and contrast experimentally.

If adequate rejection requires undocumented peak drive, stop increasing current. Consider a more suitable light, better collection geometry, reduced stray illumination, or a different acquisition strategy. The objective is repeatable inspection, not the largest possible brightness ratio.

Before ordering, complete this checklist:

• Record paired ambient-only and strobe-lit measurements using fixed camera settings.
• Confirm linear data and check individual pixels for clipping.
• Verify that the entire useful optical pulse fits inside the exposure.
• Document the required pulse duration, repetition rate, and burst sequence.
• Confirm the light-and-controller combination against documented operating limits.
• Validate the inspection across ambient changes, part variation, and production motion.

Take those measurements to the controller configurator, or contact the applications team with your timing requirements and image measurements before choosing a light-and-controller pair.