IR-Cut Filters
IR Cut-Off Filters in Machine Vision: A Practical Guide
What Is an IR Cut-Off Filter?
An IR cut-off filter (also called an IR blocking filter or heat-absorbing filter) is an optical filter that allows visible light to pass through while blocking infrared (IR) wavelengths. It acts as a barrier that stops the invisible IR radiation from reaching your camera sensor – ensuring your system "sees" only what the human eye would see.
Most IR cut-off filters are defined by their cut-off wavelength – the point at which transmission drops sharply. Common cut-off points include:
- 650nm – Blocks nearly all IR while allowing the full visible spectrum through. This is the most common choice for standard machine vision applications.
- 700nm – Allows a small amount of deep red/near-IR through. Sometimes used when you want slightly extended sensitivity without full IR exposure.
The transition from "passing" to "blocking" isn't instant – it happens over a slope. A well-designed IR cut-off filter has a steep transition edge, meaning it cuts cleanly between visible and infrared without bleeding or gradual roll-off.
Why Use an IR Cut-Off Filter in Machine Vision?
Most machine vision cameras use CMOS or CCD sensors that are inherently sensitive to a broad range of wavelengths – typically from around 350nm (UV) to 1000nm or beyond (near-IR). This extended sensitivity is a double-edged sword: it means your camera can "see" infrared light that's invisible to you, which can cause unexpected problems in your imaging.
Here's why controlling IR is so important:
Restore true-to-life color accuracy
Without an IR cut-off filter, infrared light sneaks into your red channel and throws off color balance. Reds appear washed out or shifted toward magenta. Greens may look unnaturally bright. Skin tones in medical or cosmetic inspection can appear unhealthy. An IR cut-off filter blocks this contamination, allowing your color camera to render accurate, natural colors that match what the human eye perceives.
Eliminate focus shift and image softness
Visible light and infrared light focus at slightly different points due to chromatic aberration in your lens. When IR reaches the sensor alongside visible light, the result is a subtle blur or "halo" effect – your image looks focused, but not quite sharp. This is especially problematic for high-resolution inspection tasks where fine detail matters. Blocking IR ensures all the light reaching your sensor focuses on the same plane.
Reduce unwanted interference from heat sources
Many industrial environments contain sources of infrared radiation: hot machinery, furnaces, incandescent lighting, sunlight through windows, even warm products on the line. Without an IR cut-off filter, these heat sources can flood your image with unwanted signal, washing out contrast and creating inconsistent exposure. Cutting the IR keeps your imaging stable regardless of thermal background.
Improve contrast and feature definition
In some cases, materials that look distinct under visible light appear similar in the IR spectrum (or vice versa). If you're relying on visible-light contrast for your inspection, IR contamination can muddy those differences. An IR cut-off filter ensures your camera responds only to the visible characteristics you're actually trying to measure.
Ensure consistency with human visual inspection
If your machine vision system is performing quality checks that will later be verified by human inspectors, you want the camera to "see" what humans see. IR cut-off filters align your camera's spectral response with the human eye, reducing discrepancies between automated and manual inspection results.
Common Applications for IR Cut-Off Filters
Color inspection and sorting
Any application where accurate color measurement matters – food grading, cosmetics inspection, textile quality control, print verification – benefits from IR cut-off filtering. Without it, your color data is contaminated by invisible wavelengths, leading to inconsistent or incorrect sorting decisions. With a proper IR cut-off filter, your camera's color response aligns with human perception and industry color standards.
Surface inspection under mixed lighting
Factory environments rarely have perfectly controlled lighting. Sunlight streaming through windows, incandescent indicator lights, warm conveyor components – all of these emit infrared. An IR cut-off filter isolates your camera from these variables, ensuring the only light it responds to is the visible illumination you've designed into your system.
Optical character recognition (OCR) and print inspection
Reading printed text, verifying labels, or checking package graphics requires crisp edges and accurate color. IR contamination softens edges and shifts colors, making reliable OCR more difficult. An IR cut-off filter keeps your text sharp and your colors true.
Medical and pharmaceutical inspection
Color accuracy is critical when inspecting pills, capsules, liquids, or biological samples. A shifted color balance could cause a system to misidentify a product or miss a contamination event. IR cut-off filters ensure the camera sees what a trained human inspector would see.
Food and beverage quality control
From checking the ripeness of fruit to verifying the color of baked goods or beverages, food inspection relies heavily on accurate color. IR radiation from warm production environments or the products themselves can interfere. An IR cut-off filter removes this variable.
Cosmetics and personal care
Inspecting lipstick shades, foundation colors, or packaging requires precise color matching. Even subtle shifts caused by IR contamination can result in products being incorrectly graded or rejected. IR filtering ensures your vision system meets the same standards as your human quality team.
General-purpose machine vision with color cameras
Even if your application isn't explicitly about color, using an IR cut-off filter with a color camera is good practice. It ensures predictable behavior, reduces variables, and gives you images that accurately represent the scene as humans perceive it.
IR Cut-Off Filters vs. Daylight Filters: What's the Difference?
You may encounter the term "daylight filter" or "IR/UV cut filter" – these are closely related but slightly different:
- IR cut-off filter – Blocks infrared, typically passes full visible spectrum and may allow UV through.
- UV/IR cut filter (daylight filter) – Blocks both ultraviolet and infrared, passing only the visible spectrum (roughly 400–700nm). This mimics the response of the human eye most closely.
For most machine vision applications, a UV/IR cut filter (daylight filter) is the more complete solution, as it eliminates potential interference from both ends of the spectrum. However, if your lighting is purely visible (such as LED arrays with no UV component), a simple IR cut-off filter may be sufficient.
How to Choose the Right IR Cut-Off Filter
Step 1: Confirm your camera's spectral sensitivity
Check your camera's datasheet for its spectral response curve. Most machine vision sensors are sensitive well into the near-IR range (often to 900nm or beyond). If your camera has built-in IR filtering, you may not need an external filter – but many industrial cameras ship without one, expecting you to add optical filtering appropriate to your application.
Step 2: Assess your lighting environment
Consider all sources of light in your imaging area – not just your designed illumination. Is there sunlight? Incandescent bulbs? Warm machinery nearby? The more IR sources present, the more important a high-quality IR cut-off filter becomes.
Step 3: Decide whether you also need UV blocking
If your environment includes UV sources (certain inspection lights, sunlight, some fluorescent fixtures) or if your application is color-critical, consider a combined UV/IR cut filter rather than IR-only. This gives you the cleanest visible-light-only response.
Step 4: Check the cut-off wavelength and slope
A typical IR cut-off filter cuts at around 650nm. Look for a steep transition slope – this means the filter cuts cleanly between visible and IR rather than gradually rolling off. A sharp cut gives you better control and more predictable results.
Step 5: Consider transmission efficiency
A good IR cut-off filter should have high transmission (90%+) across the visible spectrum. Low transmission means less light reaches your sensor, requiring longer exposures or brighter illumination. Check the filter's transmission curve to ensure it's not robbing you of valuable signal.
Step 6: Match the filter to your lens and mounting system
IR cut-off filters are available in threaded mounts (M25.5, M27, M30.5, M35.5, etc.), drop-in formats, and unmounted glass for custom integration. Make sure you select a size and format that fits your lens or housing. Some cameras also accept internal filter holders in the C-mount or CS-mount back-focus area.
A Few Practical Tips
- Don't assume your camera has a built-in IR filter. Many industrial and machine vision cameras ship without one – it's left to the integrator to add filtration appropriate to the application. Check your camera's documentation or test empirically by pointing it at an IR remote control; if you see the LED light up on screen, you have IR sensitivity.
- Beware of cheap filters with soft edges. Low-quality IR cut-off filters may have gradual transition slopes that allow some IR bleed-through. For critical color applications, invest in a filter with a sharp, well-defined cut-off.
- Watch for reflections. Like any optical element, IR cut-off filters can introduce surface reflections if not properly coated. Look for filters with anti-reflection (AR) coatings to maintain image quality and reduce ghosting.
- Consider filter placement. IR cut-off filters can be placed in front of the lens, behind the lens (in the back-focus area), or in a filter holder between lens elements. Each position has trade-offs for image quality and convenience – consult your system integrator or lens manufacturer for guidance.
- Test under real conditions. The best way to confirm an IR cut-off filter is working as expected is to test your system under actual production lighting conditions. Compare images with and without the filter to see the difference in color accuracy, sharpness, and contrast.
When You Might NOT Want an IR Cut-Off Filter
While IR cut-off filters are valuable in most color imaging applications, there are situations where you specifically want to capture infrared:
- Night vision or low-light imaging – IR illumination with an IR-sensitive camera can image scenes in complete darkness.
- Seeing through certain materials – Some plastics, inks, and coatings are transparent to IR but opaque to visible light. IR imaging can reveal hidden features.
- Moisture or contamination detection – Water and some organic materials absorb IR strongly, making them visible under IR imaging when they'd be invisible in the visible spectrum.
- Thermal inspection – Though true thermal imaging requires specialized cameras, near-IR imaging can reveal temperature-related differences in some materials.
In these cases, you'd want to remove any IR cut-off filter and possibly add an IR pass filter instead.
Bringing It Together
An IR cut-off filter is a small but essential component for any machine vision system using color imaging or operating in environments with significant infrared radiation. It ensures your camera sees what you intend it to see – accurate colors, sharp focus, and consistent contrast – without interference from invisible wavelengths.
For most applications, a quality IR cut-off filter (or UV/IR cut filter) paired with controlled visible LED lighting is the foundation of reliable, repeatable imaging. It's a simple addition that eliminates a whole category of potential problems.
Need help selecting the right IR cut-off filter for your application? [Explore our optical filter range →https://www.kupooptics.com/en/collections/ir-cut-off-filters] or contact us for application support.
Frequently Asked Questions
https://www.kupooptics.com/en/blogs/filter-products/mv_ir_cut_filter
What does this article cover?
This article covers IR-cut filters for machine vision: what they are, why they matter, how to select the right one, and what specifications to check before buying. An IR-cut filter is an optical shortpass filter with a cut-off wavelength between 650 nm and 720 nm. It transmits visible light (roughly 400–650 nm) and blocks near-infrared radiation (above ~700 nm). In machine vision, it is placed in the optical path between the lens and the sensor to prevent NIR light from reaching the imaging chip.
What Is an IR-Cut Filter?
An IR-cut filter is an optical shortpass filter with a cut-off wavelength between 650 nm and 720 nm. It transmits visible light (roughly 400–650 nm) and blocks near-infrared radiation (above ~700 nm). In machine vision, it is placed in the optical path between the lens and the sensor to prevent NIR light from reaching the imaging chip. IR-cut filters are also called hot mirrors because they reflect (rather than absorb) the infrared portion of the spectrum—the reflected IR goes backward, away from the sensor, while visible light passes through.
Why Do Machine Vision Systems Need an IR-Cut Filter?
Silicon-based image sensors respond to light from approximately 350 nm to 1100 nm—much broader than the human visible range (400–700 nm). Without IR filtering, NIR radiation (700–1000 nm) contaminates every pixel alongside visible light. This causes several problems: color error in RGB cameras (NIR leaks into all three color channels equally, washing out color accuracy); contrast loss in monochrome cameras (NIR from non-imaging sources adds background signal); focus shift (NIR focuses at a different plane than visible light, so lenses focused under visible illumination are soft under broadband light).
What Parameters Define an IR-Cut Filter?
Cut-off wavelength (λ_c): wavelength at 50% transmission—typically 650, 680, 700, or 720 nm. Steeper cut = better spectral separation but narrower pass band. Edge steepness: wavelength range from 80% to 20% T—steeper is better. Peak transmission: should exceed 90% in the visible pass band. Blocking depth: OD ≥ 3 (99.9% blocked) is standard for machine vision; OD ≥ 4 for color-critical work. Reflection vs. absorption: hot-mirror types reflect NIR (risk of ghost images if housing is reflective); absorptive types convert NIR to heat (better for enclosed or compact systems).
How Does the Cut-Off Wavelength Affect My Imaging System?
Moving the cut-off lower (e.g., 650 nm) means more IR is blocked but some deep red is also attenuated—this can cause color shifts in red-heavy scenes or cause red LEDs near 660–680 nm to appear dimmer. Moving the cut-off higher (e.g., 720 nm) passes more red but allows some NIR in (700–720 nm region). For most color machine vision applications, a 700 nm cut-off is a good compromise. For applications using 660 nm or 670 nm laser diodes, verify the filter's transmission at that exact wavelength before purchasing.
What Should I Check When Sourcing an IR-Cut Filter?
Request the spectral plot (transmission vs. wavelength curve), not just the nominal cut-off. Check the AOI (angle of incidence) spec—cut-off wavelength blue-shifts at non-zero angles. Verify the physical format matches your lens mount (C-mount thread pitch, S-mount, square filter holder). Confirm hard-coat vs. soft-coat: hard-coated filters are more durable and cleanable. Check if the filter has AR (anti-reflection) coating on both sides—this reduces surface reflections from ~4% per surface to <0.5%, improving overall throughput.
