Using a thermal camera effectively comes down to three things: setting it up correctly before you shoot, understanding what you’re looking at on screen, and knowing the situations where infrared imaging has blind spots. Most beginners point and shoot without adjusting key settings, which produces washed-out images and inaccurate temperature readings. A few minutes of preparation makes a dramatic difference in image quality and measurement reliability.
Let Your Camera Reach Thermal Equilibrium First
When you first power on a thermal camera, the sensor and internal electronics are adjusting to ambient temperature. If you start capturing images immediately, temperature readings can drift by several degrees. Give the camera at least five to ten minutes of warm-up time before you trust its measurements. This allows internal thermal gradients to equalize and the sensor to stabilize at its operating temperature.
If you’re moving between environments with big temperature differences, like stepping from an air-conditioned truck into summer heat, restart this warm-up period. The camera’s internal reference needs time to recalibrate to the new conditions.
Set the Right Emissivity for Your Target
Emissivity is the single most important setting on a thermal camera, and it’s the one most people ignore. Every material radiates infrared energy differently. Emissivity is measured on a scale from 0 to 1, where 1 means the surface radiates heat perfectly and 0 means it reflects everything and radiates nothing. If your emissivity setting doesn’t match the material you’re measuring, your temperature readings will be wrong.
Most organic and painted surfaces have high emissivity and are easy to measure accurately. Concrete sits around 0.92 to 0.97. PVC ranges from 0.91 to 0.93. Planed wood is about 0.90. You can leave your camera at its default setting (usually 0.95) for these materials and get reliable results.
Metals are where things get tricky. A polished aluminum surface has an emissivity of just 0.05, meaning it reflects almost all the infrared energy hitting it instead of emitting its own. Polished brass is 0.03, and polished copper is 0.05. Point a thermal camera at these surfaces without adjusting emissivity and you’ll get a reading that reflects the temperature of whatever is behind you, not the metal itself. Oxidized metals behave very differently: rusted iron jumps to 0.91 to 0.96, and oxidized copper reaches 0.65. If you’re inspecting metal surfaces, check whether they’re shiny or corroded before you trust your readings.
Choose the Right Focus Method
Thermal cameras come with different focus systems, and using the wrong one for your situation costs you both image sharpness and measurement accuracy.
- Fixed focus cameras are point-and-shoot devices that work for targets about 1.5 feet away and farther. They’re simple but low resolution, best suited for quick scans rather than detailed inspections.
- Manual focus lets you dial in on targets as close as 6 inches and still get sharp images at longer distances. This gives you the most control and generally the most accurate results, but it takes practice to get right.
- Autofocus saves time on routine scans, though you may still need to make manual corrections if the camera locks onto the wrong surface.
- Laser-assisted autofocus uses a built-in distance meter to calculate exactly how far away your target is, then adjusts the focus engine automatically. This is the fastest way to get precise images, especially when scanning multiple targets at different distances.
Regardless of focus type, always confirm the image looks sharp before saving it. An out-of-focus thermal image spreads heat signatures across adjacent pixels, which pulls temperature readings toward the average and masks the hotspots or cold spots you’re trying to find.
Adjust Level and Span for Better Images
Two controls dramatically change how useful your thermal image looks: level and span. Understanding them is the difference between a muddy grey blob and a crisp, informative thermogram.
Level sets the midpoint temperature displayed on screen. Adjusting it shifts the entire color scale up or down. If you’re looking at a wall and the interesting temperature range is around 70°F but your display is centered on 90°F, the wall will appear as a uniform color with no visible detail. Shifting the level down centers the palette on the range you care about.
Span controls how wide the temperature range is. A large span might display everything from 30°F to 130°F, which means small temperature differences get compressed into nearly identical colors. Narrowing the span increases contrast, making subtle 2 to 3 degree differences pop visually. For most inspections, you want the narrowest span that still captures all the temperatures in your scene.
These adjustments change only how the image appears on screen. They don’t alter the underlying temperature data, so you can experiment freely without losing information.
Account for Distance and Humidity
The atmosphere absorbs infrared radiation, even over short distances. Water vapor and carbon dioxide in the air each absorb a portion of the signal your camera is trying to read. At a range of just 10 meters (about 33 feet), water vapor accounts for roughly 8% signal attenuation and carbon dioxide another 6%. This means the farther you are from your target, the lower and less accurate your temperature readings become.
For close-range work indoors, atmospheric effects are minimal. But if you’re scanning rooftops, building facades, or outdoor equipment from a distance, you need to compensate. Most professional cameras have fields where you can enter the distance to target and ambient humidity so the software corrects for atmospheric losses automatically. Fill these in rather than leaving them at defaults.
Wind also matters indirectly. Moving air cools surfaces through convection, which changes the actual surface temperature you’re measuring. A pipe that reads 150°F on a calm day might read 130°F in a stiff breeze, not because your camera is wrong, but because the surface genuinely cooled. Note wind conditions when you’re comparing readings taken at different times.
Know What Thermal Cameras Cannot See Through
Glass is the most common source of confusion. Standard glass acts as a mirror for infrared wavelengths. Instead of letting infrared radiation pass through, it reflects thermal energy back toward the source and the surroundings. When you point a thermal camera at a window, you’re measuring the temperature of the glass itself, not anything on the other side. You’ll also often see a “ghost” image of yourself or nearby warm objects reflected in the glass, which can easily be mistaken for something behind the window.
This means you cannot use a standard thermal camera to inspect the inside of a building through its windows, check electrical panels behind glass doors, or evaluate anything sealed behind a glass barrier. If you need to image through a transparent surface, specialized germanium or zinc selenide windows designed for infrared transmission exist, but they’re expensive and application-specific.
Other materials that block or distort thermal readings include polished metal surfaces (which reflect infrared like mirrors), water (which absorbs infrared within millimeters of the surface), and thick insulation (which masks temperatures on the other side by design).
Practical Tips for Cleaner Results
When scanning a building envelope for air leaks or insulation gaps, work during the early morning or late evening when the temperature difference between inside and outside is greatest. A minimum difference of about 18°F (10°C) between indoor and outdoor temperatures gives you enough thermal contrast to spot problems clearly.
If you’re measuring a shiny surface and can’t adjust emissivity precisely, apply a strip of electrical tape to the surface and let it reach thermal equilibrium for a few minutes. Electrical tape has an emissivity near 0.95, giving you a reliable reference point right next to the material you’re inspecting.
Always capture a visual photo alongside your thermal image. Thermal images lack the spatial detail and context of visible light, and it’s surprisingly easy to forget exactly what you were looking at when reviewing images later. Most modern thermal cameras have a built-in visible light camera for exactly this purpose.
For cameras with sensitivity ratings (listed as NETD), a value of 40 millikelvin or lower means the sensor can distinguish temperature differences as small as 0.04°C. This level of sensitivity matters in complex scenes like dense foliage, cluttered mechanical rooms, or building inspections where the temperature anomalies you’re hunting are subtle. Lower-sensitivity cameras may miss these small differences entirely.