Focal length and field of view in thermal cameras determine how much of a scene an infrared module can observe and how many detector pixels are allocated to each object in that scene. For OEM engineers, these parameters are not only optical specifications; they affect detection range, measurement repeatability, mechanical envelope, stabilization requirements, processing load, and product cost. A lens that is too wide may cover the required area but undersample the target, while a lens that is too narrow may provide strong target detail but miss operational context.
How Does Focal Length Affect Field of View in Thermal Cameras
Focal length is the distance, usually expressed in millimeters, that defines the angular magnification of an optical system. In a simplified camera model, shorter focal lengths produce wider angular coverage, while longer focal lengths produce narrower angular coverage and higher magnification. In thermal cameras, this relationship is constrained by detector format, pixel pitch, lens material, cold-shield design for cooled systems, and the spectral band.
Field of view is the angular extent of the scene projected onto the detector. It is normally specified as horizontal, vertical, or diagonal FOV. For a rectilinear lens, the approximate relationship is:
FOV = 2 × arctan(sensor dimension / 2 × focal length)
The sensor dimension must match the FOV direction. Horizontal FOV uses detector width; vertical FOV uses detector height. For example, a 640 × 512 detector with 12 µm pixel pitch has an active width of 7.68 mm and height of 6.144 mm. With a 25 mm lens, the approximate horizontal FOV is 17.5°, and the vertical FOV is 14.0°. Changing only the lens to 50 mm narrows those values to about 8.8° and 7.0°.
The same focal length does not produce the same FOV on every module. A 25 mm lens on a 1280 × 1024 detector covers a wider angle than on a 640 × 512 detector if pixel pitch is equal. This is why focal length must be evaluated together with resolution and pitch. High-resolution cores such as the SPECTRA L12 1280×1024 LWIR can support wider coverage while preserving more pixels on target than lower-resolution formats at the same lens focal length.
Focal Length vs Field of View: What Is the Difference
Focal length is a property of the lens assembly. Field of view is the angular result of combining that lens with a specific detector. The distinction matters because OEM specifications often mix the two. A requirement such as “25 mm lens” is not equivalent to “18° horizontal FOV” unless the detector format is also fixed.
Field of view describes coverage. It answers the question: how much angular scene width does the camera see? Focal length describes optical magnification. It answers the question: how strongly does the lens project a distant object onto the focal plane? Two thermal modules can use the same focal length but deliver different FOV because their detector active areas differ.
Detector pixel pitch also changes the engineering interpretation. A 640 × 512, 12 µm LWIR detector has the same active size as a 1280 × 1024, 6 µm detector, so both can produce similar FOV with the same lens focal length. However, the 1280 format places more pixels across the same scene angle. This improves sampling, supports digital zoom, and can improve algorithmic classification, provided the optics, focus, NETD, and processing chain support the additional detail.
For cooled MWIR modules, focal length selection is also tied to cold-stop matching and f-number. Products such as the SPECTRA M06 640×512 Cooled MWIR 15μm are often used where range, sensitivity, and optical efficiency are more critical than compactness. The lens cannot be treated as a generic front accessory; it must be matched to the cooled detector and dewar assembly.
How to Calculate Pixels on Target for Thermal Camera Lens Selection
Field of view alone does not determine whether a thermal camera can detect or classify a target. The practical question is pixels on target. A wide FOV may show the full scene, but each object occupies fewer pixels. A narrow FOV increases target sampling but reduces situational coverage.
Instantaneous field of view, or IFOV, is a useful starting point. It is the angular subtense of one pixel and is commonly approximated as:
IFOV = pixel pitch / focal length
The result is in radians when pitch and focal length use the same units. A 12 µm detector with a 25 mm lens has an IFOV of 0.00048 radians, or 0.48 mrad. At 1,000 m, one pixel corresponds to about 0.48 m on the target plane, before accounting for optical blur, atmospheric effects, focus error, motion, and image processing.
For a target of known width, the approximate number of pixels across the target is:
Pixels across target = target width / ground sample size
Ground sample size increases with range. If the target width is 2 m and the pixel footprint is 0.48 m at the target distance, the target spans about four pixels. That may be adequate for detection in some scenarios but insufficient for reliable recognition or identification. The required pixel count depends on task, contrast, background clutter, algorithm design, and probability requirements.
In OEM programs, this calculation should be performed early for each operating mode. A perimeter camera for Border Security may prioritize long-range detection and therefore need a longer focal length or a continuous zoom lens. A driver-assistance or situational-awareness camera for Vehicle integration may prioritize wide coverage and low latency, accepting fewer pixels on distant targets.
Sensor characterization standards can help separate optical sampling from detector performance. EMVA 1288, maintained by the European Machine Vision Association, provides a framework for camera characterization and terminology relevant to imaging performance: [EMVA](https://www.emva.org/standards-technology/emva-1288/). While thermal camera evaluation includes additional infrared-specific parameters, consistent measurement practice is essential when comparing modules and lenses.
When to Use Wide-Angle vs Narrow-Angle Thermal Camera Lenses
Wide-angle thermal lenses are appropriate when scene coverage is more important than long-range detail. They are commonly used for navigation, obstacle awareness, robotics, close-range monitoring, and distributed sensing. A wide FOV reduces the number of cameras needed to cover a given area, simplifies alignment, and helps operators or algorithms maintain context. The trade-off is lower pixel density on each target.
Narrow-angle lenses are appropriate when the system must observe small or distant objects. They are common in border monitoring, airborne ISR, long-range surveillance, and precision inspection from a fixed standoff distance. Narrow FOV increases pixels on target and improves measurement of small thermal features, but it also increases pointing sensitivity. Small mechanical shifts, vibration, or stabilization errors become more visible in the image.
For UAV and gimbal applications, focal length selection must account for platform motion and stabilization bandwidth. A long focal length may meet the range requirement on paper but become difficult to use if vibration, jitter, or aerodynamic disturbance spreads the target over multiple pixels. In Airborne/UAV systems, the optical design, gimbal, frame rate, exposure time, and image stabilization must be treated as a combined system.
Dual-band modules add another consideration: matching fields of view across sensors. In visible-plus-thermal products such as the FUSION LV1225A 1280×1024+2560×1440, different detector sizes and lens paths must be calibrated so that image registration remains useful across the required range. Even when two channels have similar nominal FOV, parallax, distortion, focus distance, and boresight tolerance can affect fusion quality.
What Parameters Matter Besides Focal Length and Field of View
F-number is one of the most important secondary parameters. It is the ratio of focal length to entrance pupil diameter and affects how much infrared energy reaches the detector. Lower f-number optics collect more energy but can be larger, more expensive, and more difficult to manufacture. In thermal imaging, f-number also affects sensitivity, depth of field, and compatibility with detector optics.
Spectral band changes lens material and coating choices. LWIR lenses commonly use materials such as germanium or chalcogenide glass. MWIR systems may use different materials and are often paired with cooled detectors. SWIR cameras can use glass optics more similar to visible/NIR systems, but their illumination, reflection behavior, and target physics differ from thermal emission bands. A focal length value is therefore not enough to specify an optical system across LWIR, MWIR, and SWIR products.
Distortion is another practical parameter. A wide-angle lens may have barrel distortion that changes the mapping between scene angle and pixel position. This matters for measurement, tracking, image stitching, and multi-sensor registration. For AI systems, distortion should be represented in calibration data or corrected in preprocessing so that training and deployment images remain consistent.
Focus range and athermalization are also critical. Infrared lens materials and mechanical housings change with temperature, which can shift focus. A fixed-focus thermal camera may be acceptable for a controlled industrial distance but unsuitable for a vehicle or outdoor surveillance system with large temperature variation. Athermal lens design, motorized focus, or software-assisted focus control may be required.
System integration standards can influence lens and FOV choices indirectly. Networked video products may need to report imaging streams, metadata, and control interfaces consistently. ONVIF profiles are commonly referenced in IP video system integration: [ONVIF](https://www.onvif.org/profiles/). For measurement-oriented systems, temperature units and calibration traceability should also be handled consistently with recognized metrology references such as NIST SI guidance: [NIST](https://www.nist.gov/pml/owm/metric-si/si-units).
How to Choose Focal Length and Field of View for OEM Integration
A practical OEM selection process starts with geometry, not with a lens catalog. Define the required scene width, target size, operating distance, and minimum pixels on target. Then convert those requirements into FOV and IFOV targets. Only after that should the team select detector format, pixel pitch, focal length, and lens type.
The next step is to test against real scene conditions. Thermal contrast, atmospheric transmission, optics temperature, enclosure window material, and platform vibration can all reduce effective performance. A lens that satisfies the geometric calculation may still fail if the system has insufficient signal-to-noise ratio, focus stability, or mechanical rigidity.
Mechanical constraints must be evaluated at the same time. Long focal length lenses increase total track length, mass, and moment load. Wide-angle lenses can introduce larger chief ray angles, distortion, and enclosure window challenges. In compact OEM products, the correct optical choice is often the shortest focal length that still provides enough pixels on target at the required range.
For AI-enabled products such as the NEXUS LV0619B AI multi-band Ethernet/SDI, lens selection should also consider training data and algorithm behavior. Changing FOV changes object scale, background context, and motion patterns. If the deployed lens differs significantly from the lens used during dataset collection, model performance can degrade even when the image appears acceptable to a human reviewer.
In conclusion, focal length and field of view are system-level decisions. The right choice balances coverage, range, pixel sampling, sensitivity, mechanical design, calibration, and software requirements. For OEM selection, the most reliable approach is to begin with target geometry and operating environment, then match the infrared module and lens as a combined imaging subsystem.
FAQ
How do I calculate field of view for a thermal camera?
Use the detector active dimension and lens focal length: FOV = 2 × arctan(sensor dimension / 2 × focal length). Use detector width for horizontal FOV and detector height for vertical FOV. The result is usually converted from radians to degrees.
Does a longer focal length improve thermal camera range?
A longer focal length increases pixels on target at a given distance, which can improve detection, recognition, or identification. It does not automatically improve range if sensitivity, atmospheric transmission, focus stability, vibration, or target contrast are limiting factors.
What is a good field of view for thermal surveillance?
There is no universal value. Wide FOV is useful for area awareness and short-range coverage. Narrow FOV is better for long-range target detail. Surveillance systems often use multiple cameras, zoom optics, or paired wide and narrow channels to balance context and range.
Why do two thermal cameras with the same lens have different FOV?
They may have different detector sizes or pixel pitches. Field of view depends on focal length and active sensor dimension, not focal length alone. A larger active detector area produces a wider FOV with the same lens.
Should OEMs choose lens focal length before selecting the thermal module?
Usually no. The better sequence is to define range, target size, scene coverage, and pixels-on-target requirements first. Then select detector format, pixel pitch, spectral band, lens focal length, and mechanical package together.
