How to Fix Thermal Tracking Ghosting


To put it simply, thermal ghosting on walls is an issue that arises when an occupant has been in a room for a long time and then leaves. The wall will still have some lingering heat from their presence even though they are no longer there. This can make the visuals on any thermal camera completely useless as it leads to incorrect interpretations of what is happening in the room. This problem is also known as “Thermal Tracking Ghosting” because you might see something appear and “track” across the wall, but it’s not really there. It was just leftover energy! IN this article, I will discuss how to fix thermal tracking ghosting on walls.

How to Fix Thermal Tracking Ghosting

How does Thermal Tracking Ghost Happen?

The first rule of thermal tracking ghosting is that ghosts only appear after people have been in a specific place for enough time for their body heat to impact the building’s structure. In a nutshell:

But what about all those thermal tracking images posted online that don’t have anyone in them? Those ghosted figures are still real; they were just there long enough before anyone turned on their FLIR camera to create an imprint of themselves.

Using a FLIR camera and similar thermal imaging equipment, we can see these thermal tracking ghosts because the imprints left by humans stand out at lower temperatures than nearby surfaces. But even though these strange artifacts occur naturally through normal heating and cooling cycles, individuals can cause ghosting without being present in the area intentionally.

Step-wise Guide on How to Fix Thermal Tracking Ghosting on Walls:

It’s likely that if you have Thermal Tracking Ghosting on Walls, then your thermal camera will become virtually useless in low temperatures. This is a problem that can be solved using the following steps:

Step 1:  Find Out What Temp Your Camera Is Recording At

The first step to fixing this issue is to find out what temperature your camera is recording at. This means running an image analysis with your thermal camera and noting the temperature it displays as hot/cold. Most likely, what you’ll see is everything is displayed as cool (blue) or warm(red).

Step 2:  

Increase Camera Power Supply Voltage If you see everything as cool or warm, then chances are some applications require more power from your camera than it can supply. This can be fixed by increasing the voltage in your power supply.

Step 3: 

Try A Different Lens Most cameras come with two lenses, a wide-angle lens, and a telephoto lens. If you see everything as cool or warm, then try using the other lens for better results. Some cameras only have one lens, so if this is the case for your camera, then you’ll need to wait until it warms up enough to see what temperature things are really at.

Step 4: 

Decrease Camera Power Supply Voltage Once you’ve found out what temperature things are at, test different temperatures and note down which ones work well and which ones don’t work so well by creating an image analysis map of all your readings from coldest to warmest. Most likely, you’ll notice your camera works really well at a temperature just below freezing (0˚C) and doesn’t work so well when it’s warmer than -10˚C. By trial and error, you should be able to find the sweet spot that allows your camera to operate at its best in all temperatures.

Step 5: 

Finish The Job

Once everything is displaying accurately again, please put it back together with all the bits and pieces held on by tape or super glue. If any paint is still stuck around the lens area, then use acetone to remove it from around the lens; otherwise, this will block out a small part of your thermal image, leaving ghosting behind.

Precautions While Fixing Thermal Tracking Ghosting on Walls:

You must keep in mind that your Raspberry Pi camera needs to be adjusted correctly for thermal tracking. It would help if you placed it at an angle of putting out the smallest picture possible towards the target object. If you don’t do this, you will notice that some parts of your room may get calculated as much heat as they put out, and therefore several spots will show up on thermal imaging or just a blank white area.

This happens more easily if there are walls behind the target object because the thermal radiation from those walls has a different temperature than other objects in the room, causing them to show up differently on thermal imaging, which causes ghosting effects like these.

If you don’t have a direct line of sight to an object with a Raspberry Pi camera, you can put a lens in front of the Raspberry Pi to get its angle. These lenses are not that expensive on eBay or Amazon, and they make thermal imaging easier if there is no line of sight.

You need to keep in mind that you must use Thermal Track software with Raspberry Pi cameras for thermal tracking because RTSP streaming will not work properly even though it works well with regular IP cameras. You can fix this by installing an RTSP streamer, which provides both RTSP streaming and MJPEG access at the same time. That way, you can switch back and forth between one or the other.

Can I Paint Over Ghosting?

No, you can’t. The paint will not stick to the surface that has been hot-tracked. It’s like a layer of hard plastic coating on top of the wall or ceiling that wasn’t there before. You’ll end up with an awkward finish, and it won’t last for long anyway because moisture will penetrate the painted area and cause it to peel off within months. Besides, soon after one coat over ghosting, the following someone again with heat tracking sensors may expose their previous tracks leading right back to your location once more!


If you have thermal tracking ghosting on walls or ceilings, the best thing to do is find out what parts of your house are hot-tracked and see if you can move things around, paint them, or whatever. I hope this article has been beneficial for learning how to fix thermal tracking ghosting. Thank you and have a nice day!

You may read How to Melt Copper Pennies?

We will be happy to hear your thoughts

Leave a reply

DIY Quickly