Thermal Imaging: “Image Freezing”: The Myth Of Moving Too Fast

You’ve likely experienced it before. You’re moving quickly down a hallway at a residential structure fire, occasionally glancing at the thermal imager (TI) in order to keep track of adjacent rooms and accountability with your team, when you see the image momentarily freeze on the display on your TI. Why did that happen? You look at the screen for a few extra moments and everything seems to be working fine, so you continue down the hall. As you approach the bedroom at the end of the hallway, the heat conditions begin to increase and you can hear the popping and cracking of a well-seated structure fire, but with the thick smoke, you have yet to actually see it. You glance at the TI again, just to see if you can locate the fire. As you raise the TI, you can definitely see it, but it happens again – the image once again freezes on the display of the imager. What is wrong with this thing?


What are they seeing?

I am fortunate enough to be able to interact and be involved with quite a few fire departments around the world and I have a lot of conversations with firefighters from all segments of the profession. Usually, these conversations have something to do with requests for information or assistance related to thermal imaging. Several times in the past few months, the issue of image freezing has come up (although not the topic of the conversation) and each time, the firefighters’ concept of what had happened was almost completely inaccurate. What was most alarming was that most of these firefighters were “trained” to believe what they believe.

It has been just over two years since we last discussed this topic in this column (see “The Highs and Lows of Gains,” February 2010), but the myths and inaccuracies not only persist, but are being propagated under the guise of training. Here are some of the reasons I hear:

• “Our imager is really old.”

• “There was something wrong with it, so we sent it in for repair.”

• “Sometimes, you gotta smack it on the side to get it unstuck.”

• The most common – “We were just moving so fast, the TI could not keep up.”

So, why does the image freeze on the display of the thermal imager? Simple. Non-uniformity correction (NUC). See? Totally clear now, right?


NUC is the process by which the TI calibrates itself. The detector in your thermal imager is a pixel array, much like that on a digital camera. If your TI resolution is 320 x 240, then you have 76,800 pixels and each one can generate its own heat measurement. These 76,800 pixels update 30 times per second and are often exposed to widely swinging temperature scenes. These pixels also heat up themselves, from intrinsically generated heat, environmentally exposed heat and absorbed heat from what is being viewed. As the pixels heat up, they heat at different rates and therefore interpret temperatures differently from each other. The pixels must be periodically calibrated in order to eliminate this difference and this calibration is done via an internal shutter.

During the NUC process, an internal shutter closes, blocking all incoming heat and creating a common temperature reference point for all pixels to calibrate against. With the shutter closed, the pixels reset without the influence of incoming heat, but with the influence of internal heat. This ensures that the imager is displaying accurate differences in heat, even though the pixels themselves are warming up. If this NUC process did not happen, the imager would become less and less accurate and ultimately be unable to generate any image at all. This entire process typically takes less than a half-second.

Influences on NUCs

So what determines when a particular imager will perform an NUC? There are two influences on NUCs. The first influence is an “as needed by the imager” process. This can depend on the manufacturers, but it varies widely. Most NUCs are based on firmware/software algorithms that are running in the background to determine when an NUC is necessary to improve or maintain picture quality. The specifics are far too technical and vary too widely to get into here, but from the observer’s (firefighter’s) standpoint, this process will likely occur every 30 seconds to three minutes, depending on the TI manufacturer.

The second influence for NUC is in the transition of gain states. When an imager switches from high gain (ambient temperatures) to low gain (fire temperatures), the sensitivity of the pixels themselves is being electronically changed. Because of this, it is necessary to perform an NUC as a first step in the new gain state. This happens regardless of switching from low gain to high gain or from high gain back to low gain. After the confirmatory NUC at gain switch, the imager will revert to software/firmware control on an as-needed basis.

There is nothing you can do to stop this from happening and there is nothing you can do to make it any faster. The only thing you do that has any effect on NUC whatsoever is to cause the imager to change gain states. If you look at the fire and then look away, this will cause two NUCs. Otherwise, it will occur on its own.

Complicating matters is the fact that first-generation thermal imagers did not have this issue, so if you were originally trained on an older TI, you likely never saw this image freezing. Of course, if you remember, first-generation TIs were also susceptible to white-out at high temperatures and were not as rugged and durable as TIs in the market today.

The bottom line is this: it is normal and it has nothing to do with how quickly you are moving. These types of thoughts can lead to a belief that using the TI will slow you down or is inefficient, which lead many people to leave the TI on the truck where it does no good. Take the TI with you and use it. With the exception of your personal protective equipment, the TI is the next most important thing to saving your life or the life of another. n