A person-shaped patch appears on the thermal screen, but nobody seems to be standing there. Before asking who it is, take a careful look at what the camera is pointed toward.

Thermal camera reflections can put the operator or another warm source into an image of glass or a reflective surface. The picture may be striking, but its shape alone doesn’t tell you whether you are seeing a person directly, a reflection, or something else that needs checking.

This field routine helps you document the scene, test a reflection explanation, and preserve enough context for someone else to assess the result.

Why thermal camera reflections can be misleading

Thermal imaging and ordinary photography don’t show surfaces in the same way. FLIR explains that a typical thermal camera won’t show the scene beyond an ordinary glass window the way your eyes do; the window can instead show your thermal reflection. Reflective metal surfaces also require care.

That’s the key technical point for this exercise. A window that looks transparent to you should not automatically be treated as transparent in the thermal image. Start by identifying the surface in front of the camera.

If you’re still learning the display, the HauntGears guide to reading thermal images provides broader context. Here, the narrower question is whether a particular shape changes with the geometry of the shot.

First, make a normal-light record

Take a visible-light photograph of the same area when permitted. Include the full window, cabinet, appliance, or panel, along with nearby room features. A thermal close-up can make it hard to recognize where the suspicious patch sits.

Record where you and the other people are standing. A rough sketch is fine: camera at the doorway, person beside the left wall, window opposite. Add obvious warm sources such as a radiator or recently used equipment without touching or altering the property.

Do this before the team spreads out to look for an explanation. Once everyone has moved, the original arrangement may be surprisingly difficult to reconstruct.

Keep the first recording intact

If your camera can record, keep the sequence running while you describe the observation. Say what you see without naming it: “Warm shape near the lower right of the window.” Avoid “person in the window” until you know that description is justified.

If the device only saves stills, take an initial image and label the following test images in order. Note each change in camera position. Save the native files your camera provides as well as any convenient screenshots.

Use consistent settings where practical. If you change the palette, scale, or display mode, mark the change in your notes. A different-looking picture is difficult to interpret when the setup changed without a record.

Try a small, controlled angle change

From a safe, permitted position, move the camera slightly to one side while keeping the suspect surface in view. Then return to the original position. Observe whether the shape shifts, disappears, or reappears in a repeatable way.

The exercise is most useful when only the operator moves. Ask the rest of the group to remain still briefly. If several people reposition at once, you lose the ability to connect the result with a particular change.

A shape that follows changes in viewing angle is a reason to investigate reflection. It is not, by itself, a complete diagnosis. Keep the original and changed views together so you can compare them later.

Test the people who were actually there

Return to the starting arrangement. Have one person make a simple movement, such as raising an arm, while the camera stays fixed. Observe the suspected shape. Repeat with other relevant people one at a time, including the operator if the setup permits.

Choose an action you can recognize without dramatic gestures or moving toward unsafe areas. If the patch repeatedly follows a participant’s movement, record that relationship. The interesting result is the correspondence, not how human the patch looked in a single frame.

Don’t add people to an inaccessible room or lean over barriers to improve the experiment. If access limits prevent a useful check, document the limitation and move on.

Use a simple comparison record

A short observation table can keep interpretation honest:

• Original view: Where was the camera, and what did the shape look like?
• Angle check: What moved, and what changed in the picture?
• Participant check: Whose movement, if any, matched the shape?
• Return check: Could you reproduce the original view?
• Uncertainty: What could you not see or test?

For example, “Shape moved left when the operator stepped right and returned when the operator returned” is a specific observation. “Thermal camera detected an entity” goes far beyond it.

Attach the record to your session notes rather than relying on a filename such as “ghost1.” Neutral filenames make later review easier.

Don’t force every unusual image into one explanation

A reflection check answers one question. It doesn’t explain every warm patch, and an unsuccessful check doesn’t make a paranormal interpretation the remaining answer. You may still be missing information about the surface, the surroundings, or the device.

If the feature remains tied to one spot, record that outcome. Compare the visible image and note anything you can identify without disturbing the area. Ask the property host about ordinary uses of the space if that is relevant.

Be equally careful with an image that looks like a cold silhouette. The visual resemblance is a starting point for inspection, not a measurement of identity or cause.

Choose equipment that supports review

For investigations, consider how easily a camera lets you save files, identify settings, and compare observations afterward. Check the actual model documentation rather than assuming a phone attachment and a standalone unit offer the same controls.

A useful first practice session is at home with a familiar window and willing participant. Record the setup, run the angle and movement checks, and see whether another person can follow your notes. You’re learning the equipment’s behavior before a dramatic location adds distractions.

Browse HauntGears’ equipment directory when you’re ready to compare tools. Bring a clear requirement: a camera whose files and controls let you check a surprising image instead of only sharing it.

Share post