Aug. 13, 2025

214 - Thermal Imagers with Martin Veit

The player is loading ...

The world looks entirely different through a thermal camera lens, especially in a fire scenario. These devices reveal harsh temperature gradients between hot and cold surfaces, adding another dimension to how fire safety professionals understand and navigate dangerous environments.

Thermal cameras have transformed firefighting operations with astonishing effectiveness. Studies show that in smoke-filled buildings, thermal cameras have significantly improved the changes to identify victims. This technology dramatically reduces search times and increases survival chances, making it an essential tool for modern fire services around the world.

Martin Veit, who recently completed research for the Fire Protection Research Foundation, takes us deep into the science behind these life-saving devices. He explains how thermal cameras detect long-wave infrared radiation (7-14 micrometres) emitted by objects based on their temperature, creating images that reveal what smoke would otherwise conceal. The technology works because many combustion gases are relatively transparent in this part of the spectrum, giving firefighters a crucial advantage in zero-visibility conditions.

We explore the fascinating distinction between "measuring" precise temperatures (which requires understanding factors like surface emissivity and a bit of physics) and simply "observing" temperature differences (which can be sufficient for navigation and victim location). This distinction proves crucial when evaluating how thermal cameras should be tested and certified for firefighting applications.

The conversation delves into the challenges of current testing methods under NFPA standards, which sometimes yield inconsistent results that don't align with human perception of image quality. Martin's research investigates alternative approaches from the field of image processing that could provide more reliable and relevant evaluations, potentially improving both camera certification and opening doors to AI-assisted applications in firefighting.

Read the Martin's report here: https://www.nfpa.org/education-and-research/research/fire-protection-research-foundation/projects-and-reports/measuring-thermal-image-quality-for-fire-service-applications

----
The Fire Science Show is produced by the Fire Science Media in collaboration with OFR Consultants. Thank you to the podcast sponsor for their continuous support towards our mission.

00:00 - Introduction to Thermal Cameras

02:24 - Martin's Background and Research

08:35 - How Thermal Cameras Work

17:45 - Measuring vs Observing with Thermal Imaging

28:03 - Current Testing Standards

36:08 - Image Quality Assessment Methods

46:18 - Dataset Requirements and Future Applications

52:30 - Key Takeaways and Conclusion

WEBVTT

00:00:00.901 --> 00:00:02.782
Hello everybody, welcome to the FireScience show.

00:00:02.782 --> 00:00:10.935
In today's episode we will be talking about thermal cameras, and that's probably one of my most favorite devices that exist in the world.

00:00:10.935 --> 00:00:26.942
I would claim to say that if you have a fire safety engineering friend and you'd like to give them an amazing gift that they will enjoy a lot and the budget is on a little higher end and thermal camera is your way to go there.

00:00:26.942 --> 00:00:42.652
It's just fascinating to look through the world through the lens of thermal imaging, especially as soon as you start applying that into anything fire related and you see those harsh gradients between cold and hot surfaces, how stuff quickly cools or quickly heats up.

00:00:42.652 --> 00:00:47.912
It really gives another dimension to the eye of a fire safety engineer.

00:00:47.912 --> 00:00:51.890
But this episode is not just about fascination on thermal imaging.

00:00:51.890 --> 00:00:57.872
It's more about its practical side and the practical side in the life and safety.

00:00:57.872 --> 00:01:17.992
Thermal cameras are today a vital piece of equipment used by firefighters in the world to look further into the realm of fire when fighting fires, and we all understand how big advantage they can give to firefighters but also that a faulty thermal camera can be potentially life-threatening.

00:01:17.992 --> 00:01:40.036
So in this episode I've invited Martin Veit from Zag Frisbee, and Martin has just finished his report for the Fire Protection Research Foundation in which he was looking into different metrics of quality for thermal cameras thermal cameras that are actually being used by firefighters in the United States.

00:01:40.036 --> 00:01:50.509
In this episode we will dive a little bit deeper into the mathematical part of how to test different images and what does it mean to compare the image quality.

00:01:50.509 --> 00:02:13.492
But before we reach that point, we have a long discussion about the thermal cameras how do they work, why measuring temperature with cameras is actually quite difficult and why just observing the temperature gradient or differences in a fixed location is not that hard, what you can do with them, how do we test them and how can we improve it for the future.

00:02:13.492 --> 00:02:16.606
So let's not prolong this unnecessarily.

00:02:16.606 --> 00:02:19.233
Let's be in danger and jump into the episode.

00:02:23.700 --> 00:02:25.268
Welcome to the Firesize Show.

00:02:25.268 --> 00:02:28.801
My name is Wojciech Wigrzyński and I will be your host.

00:02:28.801 --> 00:02:58.240
The FireSense Show is into its third year of continued support from its sponsor, ofar Consultants, who are an independent, multi-award winning fire engineering consultancy with a reputation for delivering innovative, safety-driven solutions.

00:02:58.240 --> 00:03:12.044
As the UK-leading independent fire risk consultancy, ofar's globally established team have developed a reputation for preeminent fire engineering expertise, with colleagues working across the world to help protect people, property and the plant.

00:03:12.044 --> 00:03:18.584
Established in the UK in 2016 as a startup business by two highly experienced fire engineering consultants.

00:03:18.584 --> 00:03:28.149
The business continues to grow at a phenomenal rate, with offices across the country in eight locations, from Edinburgh to Bath, and plans for future expansions.

00:03:28.149 --> 00:03:36.391
If you're keen to find out more or join OFR Consultants during this exciting period of growth, visit their website at ofrconsultantscom.

00:03:37.681 --> 00:03:38.824
And now back to the episode.

00:03:38.824 --> 00:03:46.271
Hello everybody, I am joined today by Veit from from Frisbee at ZAG in Slovenia.

00:03:46.271 --> 00:03:48.085
Hey, martin, good to have you in the podcast.

00:03:48.085 --> 00:03:51.406
Hello, thanks for having me and thanks for joining me.

00:03:51.406 --> 00:04:07.765
We have an interesting topic here to discuss and that is your very recent report for the Fire Protection Research Foundation at the NFPA, which is on measuring thermal image quality for fire service applications.

00:04:07.765 --> 00:04:28.791
I have not had a thermal imaging episode yet, but I think it is quite an interesting topic that actually connects engineers, firefighters, practitioners Come on, everyone likes thermal cameras in fire safety, so perhaps let's start with your background and how you ended up in this project, right?

00:04:29.922 --> 00:04:40.971
So for my background, I'm actually not a fire safety engineer, I'm a civil engineer and then I just stumbled into the field and I guess the first entry into the field was really when I was, I guess, cold called.

00:04:40.971 --> 00:04:53.851
So a guy from my university, when he went into industry, recommended me to a partner in a company called Vibraxenernen, kenneth Håkvar Jensen, and then he asked if I wanted to work in the company as a student helper.

00:04:53.851 --> 00:05:08.324
So I worked there for half a year, which was my first entry really into fire protection engineering, and I worked on multiple different things, so fire strategy, but also I worked a bit on documentation and programming and I was supposed to stay.

00:05:08.324 --> 00:05:11.516
But then I was offered a PhD.

00:05:11.516 --> 00:05:16.942
That didn't start immediately at Aalbe University, but I was supposed to start on that one Eventually.

00:05:16.942 --> 00:05:19.846
It took a bit too long for me, it didn't go as planned.

00:05:19.846 --> 00:05:24.843
But then Grunde Jomaas, my current supervisor, reached out to me on LinkedIn.

00:05:24.843 --> 00:05:27.086
But then Grunde Jomaas, my current supervisor, reached out to me on LinkedIn.

00:05:27.086 --> 00:05:28.786
I guess I liked LinkedIn posts.

00:05:28.927 --> 00:05:32.410
And then he wrote to me in Danish, I believe on LinkedIn.

00:05:32.410 --> 00:05:35.372
So I was very surprised and offered me a PhD in Slovenia.

00:05:35.372 --> 00:05:40.237
So I thought let's go back to fire protection engineering and then take an adventure.

00:05:42.160 --> 00:05:42.800
Go to Slovenia see what brings.

00:05:42.800 --> 00:05:46.146
So there are actually positive things from Grundy's LinkedIn activity.

00:05:46.146 --> 00:05:46.827
Good to know.

00:05:47.149 --> 00:05:49.994
Absolutely Exactly.

00:05:50.173 --> 00:05:50.375
Yeah.

00:05:50.375 --> 00:05:53.247
And how did you end up studying thermal cameras?

00:05:53.829 --> 00:05:54.089
Right.

00:05:54.089 --> 00:05:56.949
So I didn't have a background in fire protection engineering.

00:05:56.949 --> 00:06:04.134
So more or less the first half a year to a year was me figuring out what am I going to do in the PhD.

00:06:04.134 --> 00:06:10.867
And I think I ended up after a meeting with Andrea Licardini, another one of my supervisors.

00:06:10.867 --> 00:06:27.966
We had a meeting and then at some point we discussed okay, maybe it could be interesting to look into gas phase measurements, and so I started reading a lot, found some papers, and then I found a paper that used thermal imaging cameras to then characterize the flow field after pool fire, and so I looked into that a bit.

00:06:28.266 --> 00:06:36.552
But then I stumbled into the fact that thermal imaging cameras normally have pretty low resolution, both spatially and also the frame rate.

00:06:36.552 --> 00:06:41.312
So I thought, okay, how can we accommodate this problem, this issue, this?

00:06:41.312 --> 00:06:59.526
So I looked into how to improve resolution of videos, so both the spatial and temporal resolution, and I recently presented a paper on that in Greece, using some machine learning to then enhance video footage, specifically of thermal videos.

00:06:59.526 --> 00:07:18.651
And then at some point during this whole process, I think he received an email about something called student project initiatives from the FPRF, so the Fire Protection Research Foundation, and we looked into the projects that they had and one of them was specifically on measuring the thermal image quality.

00:07:18.651 --> 00:07:28.274
So I think from the beginning to the end, from my background and then ending up to where I am now, has been a big coincidence, more or less, into this project also.

00:07:28.860 --> 00:07:30.447
Welcome to World of Fire Science.

00:07:30.447 --> 00:07:32.362
I guess it's a common story.

00:07:32.362 --> 00:07:43.649
I guess nine out of ten people, I think so We'll have a similar life story about how they ended up in the current location where they are working or the topics that they are dealing with.

00:07:43.649 --> 00:07:45.406
Good, so a nice coincidence.

00:07:45.406 --> 00:07:54.779
I remember talking with you some time ago I think it was the conference in Slovenia about your ideas on measuring gas flow fields and using observation.

00:07:55.081 --> 00:08:02.334
It's kind of a holy grail in fire science to be able to incorporate more of optical measurements.

00:08:02.334 --> 00:08:21.286
They are fast, they are clean, they are easy Air quotes were shown while saying easy, because they're not very easy, but they look easy and it is my strong belief that you can get much more from recording fires and observing fires than you think you could get.

00:08:21.286 --> 00:08:31.509
But the technology that you're investigating here is not specifically meant to support researchers in their experimental endeavors.

00:08:31.509 --> 00:08:35.049
It's a technology used every day by firefighters, I presume.

00:08:35.049 --> 00:08:44.687
Can you tell me more about the types of cameras and the technology and how it is currently overwhelmingly used in the firefighting profession?

00:08:45.129 --> 00:08:46.562
Yes, absolutely so.

00:08:46.562 --> 00:08:48.899
Firefighters use this technology.

00:08:48.899 --> 00:09:01.653
So the thermal imaging camera or thermal imager, also called a TIC as a short name, and they use this for a lot of different of their duties when doing structural firefighting.

00:09:01.653 --> 00:09:17.110
So going into a building and then trying to navigate the building because you might have a building with a fire, you have a lot of smoke can be very difficult to see with your eyes, and then trying to navigate the building, much less identify a person lying on the floor if you can't see anything.

00:09:17.110 --> 00:09:28.249
So there was this nationwide study in the US I believe, where they looked into the efficacy of thermal imaging cameras and the way firefighters use them.

00:09:28.249 --> 00:09:40.754
So what they found was that if a firefighter goes into a building with a lot of smoke, they I think it was 60% of the times they couldn't identify a person without the thermal imager.

00:09:40.754 --> 00:09:44.390
But with the thermal imager they could identify the person 99% of the time.

00:09:44.390 --> 00:09:49.551
Also, it reduces a lot of the time it takes to locate a person.

00:09:49.551 --> 00:09:51.120
Also, navigate your way out of the building.

00:09:51.764 --> 00:09:55.599
So is it something that you could call, now, a standard piece of equipment?

00:09:55.599 --> 00:09:57.583
I wonder how it is around the world.

00:09:57.583 --> 00:10:03.083
I think in Poland the fire brigades are very strongly equipped with thermal imaging cameras.

00:10:03.445 --> 00:10:03.664
Right.

00:10:03.664 --> 00:10:04.908
So that is also my belief.

00:10:04.908 --> 00:10:23.250
So maybe in the 2000s, I think, when they started to route them out to firefighters and then before that, not so much because they were big, they were bulky, but then, when the technology then advanced, you can have a smaller thermal imaging camera, it's easier to use for the firefighter, and so on.

00:10:23.250 --> 00:10:43.668
So for now at least, also in the US, which was the main focus of the project, because it was an American project, it seems to be very, very common in the fire service to use this thermal imaging, and also I talked to some people in Europe as well which, as you also say, very common to use because it's such an effective equipment.

00:10:44.630 --> 00:10:52.471
I wonder when it's going to be equipped as a part of, you know, the helmet and the body kit itself.

00:10:52.471 --> 00:11:07.100
I can imagine this kind of device to be integrated with the helmet and the visor in the helmet to actually present some augmented reality, thermal imaging, over what the firefighter directly sees, perhaps with some clever eye tracking.

00:11:07.100 --> 00:11:14.254
I think every single component of technology that would be necessary to do such a thing already exists.

00:11:14.254 --> 00:11:18.652
It's just about you know, scale and being able to deliver that.

00:11:18.652 --> 00:11:25.269
So I'm absolutely convinced this is an element of infrastructure that will be a part of the future.

00:11:25.750 --> 00:11:33.581
And I remember when I rejoined the ITB 15 years ago, we had this massive, bulky thermal camera.

00:11:33.581 --> 00:11:35.826
It was like a beamer size, you know.

00:11:35.826 --> 00:11:47.241
You had to have a whole bag with that, and now a thermal camera that's significantly, significantly better in technology is of the size of my, of my iphone.

00:11:47.241 --> 00:11:47.823
That that's it.

00:11:47.823 --> 00:11:48.706
That's that.

00:11:48.706 --> 00:11:51.596
That's the progress in this technology you mentioned.

00:11:51.596 --> 00:11:53.484
They have low resolution, low frame rate.

00:11:53.484 --> 00:11:58.964
So how does that compare to devices that people are used to like cameras in their phones, if?

00:11:58.984 --> 00:12:05.931
you have a standard iphone, not necessarily the newest one, but you'll have full hd images right, so you can have very high resolution.

00:12:05.931 --> 00:12:12.346
If you take an image, you can easily discern details, you see a lot of colors, it's very clear, very visible.

00:12:12.346 --> 00:12:17.628
But then for the thermal imaging camera it has a resolution much lower than that.

00:12:17.628 --> 00:12:33.373
If you have like a, if you're going to have very expensive cameras that will have a high resolution, also high frame rate, but the ones that are more affordable, also to the fire service, will typically have a resolution of 240 times 320.

00:12:33.754 --> 00:12:42.380
That's like the common one I think my first phone had a camera of 240 times 320 and it's a very low resolution actually.

00:12:42.380 --> 00:12:44.504
Uh, what about frame rates?

00:12:44.504 --> 00:12:48.067
Are we talking about like single image per second?

00:12:48.067 --> 00:12:50.370
Are we talking something like a movie?

00:12:50.370 --> 00:12:51.552
24, 30 hertz?

00:12:51.552 --> 00:12:53.835
How do they operate in frame rate?

00:12:56.820 --> 00:12:57.163
To a large degree.

00:12:57.163 --> 00:13:09.139
It's actually pretty good for what you need to do if you want to operate a thermal imaging camera and then go into a building because you don't want it to lag too much or you have too much time in between individual frames because that might make it very difficult to navigate.

00:13:09.139 --> 00:13:23.827
So the ones that are nfpa approved so the ones that are very common in the us, typically have a frame rate of 30 hertz or even as high as 60 hertz, which is pretty good for a thermal imaging camera and an fpa approved.

00:13:23.888 --> 00:13:25.231
Is there a standard they refer to?

00:13:25.231 --> 00:13:28.009
Is there something that defines what they should do?

00:13:28.009 --> 00:13:30.596
Perhaps that's something we should refer the listeners to.

00:13:31.078 --> 00:13:50.889
Right, so the standards that goes into certifying these thermal imaging cameras in the US, so the NFPA standard, that's the NFPA-1801, which is currently being consolidated into 1930, I believe it is which also collects a few other standards into one big standard.

00:13:50.889 --> 00:13:59.405
But that's the standard that goes into testing and what this thermal imaging camera should be able to withstand.

00:13:59.446 --> 00:14:10.434
And also to do and in terms of what they see, I know also in my thermal cameras there's this annoying parameter which is the temperature range which I have to set.

00:14:10.434 --> 00:14:22.155
So those devices used by firefighters, do they operate in like whole spectrum of fire temperatures up to, like I don't know, 1200s, or they are narrowed down to a few hundred degrees?

00:14:22.155 --> 00:14:26.350
What's the target temperature range on those devices?

00:14:26.940 --> 00:14:29.188
So actually, the different temperature ranges.

00:14:29.188 --> 00:14:34.285
So it sort of depends from camera to camera and the NFPA-approved ones.

00:14:34.285 --> 00:14:38.466
But I believe there must be like a minimum range in the NFPA standard.

00:14:38.466 --> 00:14:49.701
For example, some of them go from minus 20 up to 550, and then some goes from minus 40 up to 550, and then some goes from minus 40 up to 1100 degrees Celsius, so it can range.

00:14:49.701 --> 00:14:54.212
There's a big range that they can visualize and observe.

00:14:54.860 --> 00:14:57.609
But it's not like they cap at 100 degrees, right.

00:14:57.870 --> 00:14:58.149
No, no, no.

00:14:59.240 --> 00:15:00.724
And also to get utility.

00:15:00.724 --> 00:15:04.972
You don't need it to be beyond the fire range, right, no right?

00:15:05.514 --> 00:15:07.979
to get utility, you don't need it to be beyond the fire range, right?

00:15:07.979 --> 00:15:11.687
No, right, I mean, you just need it to be able to operate within the expected conditions that you want to go into.

00:15:12.240 --> 00:15:16.692
And, if you think about it, I guess the expected condition is not necessarily a fully flashed over fire.

00:15:16.692 --> 00:15:19.909
Like what kind of information does a firefighter get from a camera?

00:15:19.909 --> 00:15:26.634
Right, it's like if you have a fully flashed over fire in front of your face, you don't need to have a thermal camera to confirm that Exactly.

00:15:26.634 --> 00:15:32.873
Okay, let's move on into how thermal cameras work.

00:15:32.873 --> 00:15:36.899
So perhaps there's a bit of interesting technical knowledge.

00:15:36.899 --> 00:15:52.047
So if you could briefly explain to me why my camera in a phone does not capture the temperatures and why this specific piece of equipment does, so we can go into some of the basics.

00:15:52.159 --> 00:15:54.426
I think if you want to go into the diesel and stuff.

00:15:54.446 --> 00:15:56.230
It might be like a full episode in itself.

00:15:56.230 --> 00:16:05.379
But I guess the idea is that you have different spectrums, electromagnetic spectrum, different regions that you want to observe.

00:16:05.379 --> 00:16:15.071
So you have the visible spectrum, which is 380 up to 780 nanometers, but then the infrared is also partitioned into different regions.

00:16:15.071 --> 00:16:25.254
But for long wave infrared cameras the range is 7 microns or 7 micrometers up to 14 approximately.

00:16:25.254 --> 00:16:27.908
So it's a very different range that they observe.

00:16:27.908 --> 00:16:33.769
So we have larger wavelengths, so the individual detectors also needs to be bigger.

00:16:33.769 --> 00:16:40.607
So if you have a standard digital camera, for example, it has a much higher resolution because also it's a different technology.

00:16:40.607 --> 00:16:58.986
So there's also been a lot of research in that specific field, compared to thermal cameras, which is normally used by the military, for example, if you want to go into building inspections, fire service but there's a lot of more utility necessarily for this common person with a camera.

00:16:58.986 --> 00:17:09.627
So the main thing is also the technology used, which is why you lower frame rates, lower resolution, but that's due to some of the physics involved.

00:17:11.102 --> 00:17:13.750
How can the thermal camera distinguish between temperatures?

00:17:13.750 --> 00:17:14.151
Actually?

00:17:14.151 --> 00:17:17.328
How does it know a thing is at a low temperature?

00:17:17.328 --> 00:17:19.579
How does it know that the thing is at a high temperature?

00:17:19.941 --> 00:17:52.111
So thermal imaging cameras works on the principle that if you have any objects in front of you, it will emit a certain amount of radiation Right, and then if you have an object that's of higher temperature, it's going to emit more radiation, which is then detected by this thermal imaging camera, meaning that you can discern two different objects of two different temperatures based on the one based, because they emit different amounts of radiation that is then detected by the detector in the thermal imaging camera.

00:17:52.720 --> 00:17:59.193
Okay, I can understand how we can observe walls and solid objects with thermal cameras.

00:17:59.193 --> 00:18:03.921
What about gases like in observing flames and smoke, etc.

00:18:03.921 --> 00:18:06.141
Because I know it's not a trivial thing, about gases like in observing flames and smoke, etc.

00:18:06.141 --> 00:18:08.162
Because I know it's not a trivial thing.

00:18:08.162 --> 00:18:10.701
So how much smoke do you see on a thermal camera?

00:18:10.701 --> 00:18:12.522
How much flame do you see in a thermal camera?

00:18:12.522 --> 00:18:13.462
What do you exactly see?

00:18:14.042 --> 00:18:15.663
You have to be right, it's not trivial.

00:18:15.663 --> 00:18:23.664
It's a lot easier to detect a wall, an object, something, especially if you want to have accurate gas phase temperatures off a flame.

00:18:23.664 --> 00:18:24.826
Very difficult to do.

00:18:24.826 --> 00:18:57.526
But I guess the main principle and why this thermal imaging camera is of utility to a firefighter is because that they operate in the range of 7 to 14 microns, because if you look at some of the combustion gases that is present during a fire, you know If you look at the absorbance, they have a large range where they have very low or no absorbance in that specific range, meaning that they're not going to obscure the image that you're trying to see.

00:18:57.609 --> 00:19:01.781
So you can see true smoke because of the gases' absorbance.

00:19:01.781 --> 00:19:08.303
So, for example, I used a thermal camera, for example, to make some videos of a small pool fire.

00:19:08.303 --> 00:19:13.833
I can see that because it still has radiation, blackbody radiation, so I can still see the flame.

00:19:13.833 --> 00:19:40.803
But if I want to visualize the combustion gases and it's not necessarily super, they don't have a lot of absorbance in the specific region or the spectrum of the longwave infrared camera then I can go down and use a midwave infrared camera so I can visualize the combustion gases more clearly, meaning that I can actually see it fairly clearly, even though it's not necessarily super visible to my eye.

00:19:40.803 --> 00:19:50.751
So also flames, for example, I used a midwave infrared camera to then make videos of a pool fire and then you can also see.

00:19:50.751 --> 00:20:03.699
So it was a visible flame, so it was a heptane, so it also has some soot, but then you can also very clearly see the flame and different flame structures and then also some of the combustion gases even above the flame.

00:20:04.299 --> 00:20:05.282
Can you define mid-wave?

00:20:05.282 --> 00:20:06.693
What spectrum would that be?

00:20:06.693 --> 00:20:08.377
Is it below seven?

00:20:09.059 --> 00:20:12.799
It is, so I guess it also depends on who you ask.

00:20:12.799 --> 00:20:15.239
I've seen different things one to five or three to five.

00:20:15.239 --> 00:20:19.301
I think the one that I would choose is probably three to five microns.

00:20:19.990 --> 00:20:35.981
So in essence, the ones that would be used by the fire service let's say seven to 14 default range, basically the stuff that the fire produces in abundance CO, co2, is transparent to those radiations.

00:20:35.981 --> 00:20:48.499
Therefore you can see better through them because you're not obscured by a large amount of radiation produced by those hot gases and at the same time it makes it difficult to observe those structures because basically they're transparent as well.

00:20:48.499 --> 00:20:53.762
So it's kind of a blessing and an issue depends on on on how you look on the problem.

00:20:53.762 --> 00:21:00.182
I I had an interesting episode on on observing through flames with matt hayler from from this.

00:21:00.182 --> 00:21:16.757
So yeah, we've been talking about blue light technology, so they've also found a way to to go very near uv radiation, where UV radiation, where the flame is almost transparent to that particular blue light wavelength and you can see through flames and, yes, it actually works like that.

00:21:16.757 --> 00:21:22.423
So, in essence, a very similar principle in here.

00:21:23.349 --> 00:21:25.837
So how much you need to know to?

00:21:25.837 --> 00:21:29.820
So let's imagine you're taking a picture with a thermal camera.

00:21:29.820 --> 00:21:31.856
It gives you some output.

00:21:31.856 --> 00:21:33.276
It's a simple device.

00:21:33.276 --> 00:21:37.361
It looks on photons, accumulates them, gives you an outcome.

00:21:37.361 --> 00:21:39.758
But what exactly?

00:21:39.758 --> 00:21:49.075
You have to know about the object you're observing to understand the values that you are seeing on your screen, for example, the emissivity of the surface that you are seeing on your screen.

00:21:49.075 --> 00:21:51.853
For example, the emissivity of the surface that you're observing with thermal camera.

00:21:51.853 --> 00:21:56.653
Do you need to know the emissivity, the reflectance of the surface that you're measuring?

00:21:57.315 --> 00:22:04.780
As you mentioned, there's a few different things that you need to know, but I guess this also depends on the purpose of what you want to do.

00:22:04.780 --> 00:22:07.413
Do you want to have accurate gas-based measurements?

00:22:07.413 --> 00:22:12.455
Do you want to have accurate measurements of the temperature of some solid object, which might be very valuable?

00:22:12.455 --> 00:22:27.255
But if you don't need that, if you just want to, let's say, identify hotspots, identify heat losses of a building, for example, which is also a common use of thermal cameras, you can do that without getting all of the different specifications correct.

00:22:27.255 --> 00:22:33.298
So, for example, the emissivity, distance to the object and so on, and also ambient temperature conditions.

00:22:33.298 --> 00:22:45.277
If you need to have accurate measurements of some phenomena, let's say gas phase or sonded phase, then you need a bit more information on what to do correctly, for example the emissivity.

00:22:45.277 --> 00:22:54.698
But as far as I know, the getting accurate gas phase measurements of a flame and then fully resolve it is very difficult.

00:22:54.698 --> 00:22:56.577
It's not so straightforward.

00:22:57.059 --> 00:23:03.994
Well, technically, if you take a camera and point it towards the flame, it will tell you like 600 degrees.

00:23:03.994 --> 00:23:08.776
It's not a flame temperature of any way, so it must be incorrect.

00:23:08.776 --> 00:23:16.996
But then I, a long time before COVID, we were doing some facades within pre-rule and we were observing.

00:23:16.996 --> 00:23:23.061
We had a steel facade, one that you actually the next generation of which is sold in the summer school.

00:23:23.061 --> 00:23:28.836
We had a prototype of that Long, long time ago and we had the steel plate.

00:23:28.930 --> 00:23:34.619
There was a fire in the cavity in the steel plate and basically we were taking pictures of this facade.

00:23:34.619 --> 00:23:53.118
It was giving us a very weird numbers and we had a black spray paint which was meant for chimneys, like literally black, very dark, very matte spray paint, and we were painting on that steel facade and the locations in which we painted it it's very black.

00:23:53.118 --> 00:24:01.038
The temperature measurements in that location were like almost one degree accurate with the thermocouple measurement in the same location.

00:24:01.038 --> 00:24:18.086
So because we, like you, could very well approximate emissivity of a matte black paint, which is very, very high, whereas for the steel surface that is undergoing heating and a lot of crystal formation transitions, it changes colors, it bends a little bit, it reflects less.

00:24:18.086 --> 00:24:18.569
In this location.

00:24:18.569 --> 00:24:20.576
It was a mess.

00:24:20.576 --> 00:24:27.604
So that was a moment when I realized I really need good control over what I'm measuring.

00:24:27.604 --> 00:24:35.678
I think the word measure in here is the key, because if you want to measure, you have to put in the effort to measure.

00:24:36.380 --> 00:24:50.175
If you're just observing, here you go, you're welcome to to see the range I think it's a great distinction to have either measure or observe, because for a lot of different things that you want to do with a thermal imaging camera, observations are just fine, right.

00:24:50.175 --> 00:24:54.356
You don't need to have very accurate temperatures, you just need to discern some difference.

00:24:54.356 --> 00:24:56.184
Maybe it's hot, maybe it's cold.

00:24:56.184 --> 00:24:57.691
If it's heat losses or whatever for building.

00:24:57.691 --> 00:25:07.393
If you want to measure, you have to take a lot more care into what you're doing, control the emissaries and do one of the steps to get accurate temperatures if that's what you want to do.

00:25:07.393 --> 00:25:18.296
But I think both observation and measurement has a place in fire safety science and what we used to do with it, especially if you're trying to do smoke and fire phenomena.

00:25:19.009 --> 00:25:21.098
There is this thing called optical thickness.

00:25:21.098 --> 00:25:26.900
Basically, the emissivity of a smoke layer is not just a physical function of the suit inside.

00:25:26.900 --> 00:25:33.816
It's about how deep the layer is until you reach the optical thickness and then it looks black.

00:25:33.816 --> 00:25:53.256
And if you don't know how thick the layer is, you cannot even tell how much smoke is in the layer, because you don't know if it's like very thin but very optically thick, or it's just a little smoke spread over 10 meters which will look to you in a similar way from the emissivity of a layer standpoint.

00:25:53.256 --> 00:25:56.521
So there's a lot of cave-ins.

00:25:56.722 --> 00:26:13.509
So in the end in fires I don't think the firefighters should trust the number they see on the camera, but I think they are very well advised to trust if something is hot or not in relationship to the surroundings.

00:26:13.509 --> 00:26:21.944
If I had to give a guidance to my firefighting colleagues, who definitely are professionals in using those devices they train.

00:26:21.944 --> 00:26:40.540
But my observation as a researcher using tools like that is that wow, so hard to get an accurate measurement, but it's, it's super easy to just, you know, get an overview absolutely yeah, and I think that's also valuable to know as a firefighter, like what to do with this should I trust the temperature or not?

00:26:40.840 --> 00:26:42.613
because I also heard from a firefighter.

00:26:42.613 --> 00:26:48.452
It was in europe, but he said that they didn't have so much money to get thermal imaging cameras.

00:26:48.452 --> 00:27:06.753
So they had, I think, one good thermal imaging camera, which was for the chief, I believe and then they had cheaper thermal imaging cameras still very useful because, as you say, you can still observe information and get a lot of valuable information even though you don't have the highest resolution or you haven't, you don't have the highest frame rate at all.

00:27:06.753 --> 00:27:08.439
Still are useful too, even though you don't have the highest frame rate at all.

00:27:08.439 --> 00:27:11.750
Still are useful too, even though you don't have the newest of the new or the best of the best.

00:27:12.430 --> 00:27:16.622
Are there big differences between particular cameras?

00:27:16.622 --> 00:27:22.819
How sensitive are they to the radiation of hot objects or gases in fires?

00:27:22.819 --> 00:27:31.640
Anything like any practical differences, or you can qualify all of them as, let's say, useful for fire and that's it.

00:27:31.640 --> 00:27:33.323
No, no big differences among them.

00:27:34.044 --> 00:27:34.284
Right.

00:27:34.284 --> 00:27:46.540
Different cameras of course, made by different manufacturers, will also have differences in quality service.

00:27:46.540 --> 00:27:50.210
The ones that are certified by the NFPA standard.

00:27:50.210 --> 00:28:05.604
They primarily use two different detector technologies, so one called vanadium oxide and one called amorphous silicon, and I think at some point I believe VOX of vanadium oxide was more prevalent on the market.

00:28:05.604 --> 00:28:16.714
But I think amorphous silicon as a detector technology also caught up and is also now part of the majority on the NFPA certified thermal imaging camera list.

00:28:16.714 --> 00:28:26.978
So there's definitely different materials used and they also have different material properties that also make some slight changes for the thermal imaging camera.

00:28:26.978 --> 00:28:33.079
But I think to a large degree thermal imaging cameras that are then certified by the NFPA.

00:28:33.079 --> 00:28:34.663
They have to go through the same test.

00:28:34.663 --> 00:28:41.920
So even though it's a difference, they take the technology.

00:28:41.920 --> 00:28:43.789
It should also still be very much of a utility to the fire service.

00:28:44.250 --> 00:28:50.903
And Hjulje, I guess, get to the point of the research grant and the research question that you were involved in.

00:28:50.903 --> 00:29:12.750
How do you assess the quality of those cameras for fire applications in order to verify that when the new sensor comes to the market or a new product is introduced to the market and wants to get the certification, the certification?

00:29:12.750 --> 00:29:15.336
So let's perhaps start with how those devices would be tested in accordance to NFPA or for the purpose of certification.

00:29:15.336 --> 00:29:20.315
What exactly is being assessed when you certify a thermal camera?

00:29:20.315 --> 00:29:21.176
What you're looking into?

00:29:21.597 --> 00:29:42.000
So currently the NFPA standard that I mentioned before, the 1801, or the new one that it it's been consolidated into it goes through a bunch of tests, for durability, for example, and then the thermal image quality is then assessed with something called the image recognition test and image image recognition test.

00:29:42.000 --> 00:29:56.125
Yes, so this is then based on some work from dr francine amon back in the early 2000s, where they tried to look into different things or different ways to quantify the thermal image quality.

00:29:56.931 --> 00:30:08.015
And they looked into a bunch of different ones, for example effective temperature range, non-uniformity, spatial frequency response and the ones incorporated into the standard.

00:30:08.075 --> 00:30:25.644
Now is, the main test would be this spatial resolution test or the image recognition test, and essentially you have a bit of math that you have to do so they use something called a contrast transfer function to estimate this spatial resolution.

00:30:25.644 --> 00:30:52.875
So for this you need a blackbody target and then the test is actually a bit cumbersome because you have this thermal image that you want to certify and then to assess the quality of the spatial resolution, then you take an image of the display and then extract some region of interest and then do some math to then see okay, how well did it respond to the test.

00:30:52.875 --> 00:31:01.377
So, speaking with the technical committee, what I was told was that, first off, the test is a bit tedious to perform.

00:31:01.377 --> 00:31:16.733
It also has some consistency issues, so if you repeat it you might not get exactly the same value, also because you have this extra element of a camera taking an image off the display, and then also the image quality.

00:31:16.733 --> 00:31:22.964
If you have a human and the test side to side, they might also not agree with the image quality.

00:31:23.349 --> 00:31:30.463
So one might say okay if I said this left picture is better than the right one, the test might say otherwise.

00:31:30.463 --> 00:31:35.019
They could still both pass, but they have some consistency issues at the moment.

00:31:35.019 --> 00:31:50.397
So the whole idea with the project was to look into how can we accommodate or perhaps improve the current framework being performed to assess the thermal image quality, to make it more consistent and also to have it better aligned with what a human might score the image.

00:31:51.292 --> 00:31:57.421
Yeah, but it's not like a relative comparison between the cameras that are certified.

00:31:57.421 --> 00:31:59.617
It's more about meeting a threshold, right?

00:31:59.617 --> 00:32:07.162
Exactly, yeah, Okay, and the thing that camera is shooting do I understand?

00:32:07.162 --> 00:32:10.760
You called it the display, but it must be a hot object, right?

00:32:10.760 --> 00:32:12.978
And it's controlled by the NFPA.

00:32:12.978 --> 00:32:15.999
What pictures of what are you taking during the test?

00:32:16.769 --> 00:32:19.817
So what I meant before was that the thermal imaging camera.

00:32:19.817 --> 00:32:31.096
It takes an image, so this thermal image, of something called a stencil pattern, which is essentially just a target with a specific pattern, with a specific emissivity.

00:32:31.096 --> 00:32:36.676
But the image that you take from the thermal imaging camera is not the one that is being assessed.

00:32:36.676 --> 00:32:47.537
You have an extra step, meaning that you have a camera on the display of the thermal imaging camera that is then taking an image of that display and then from that you extract a region of interest.

00:32:47.537 --> 00:32:49.634
So you can imagine there's also some things with reflections that display and then from that you extract a region of interest.

00:32:49.634 --> 00:32:53.605
So you can imagine there's also some things with reflections that you have to take into account because you have the display.

00:32:53.605 --> 00:32:59.281
You don't want extra reflections that could change the score of the ah, I understand.

00:32:59.342 --> 00:33:09.438
So it's not just assessing the quality of the image taken by the camera, it's about assessing how you can view an object through the camera.

00:33:09.609 --> 00:33:33.003
So it also includes the assessment of the viewfinder, of the screen of the camera, in a way, exactly, and I believe the argument for this was that you have a person also looking at the display, so they want to take it from the very image that you're trying to take all the way up to the display and then treat that as a black box and then trying to assess the quality of this black box that you're trying to take all the way up to the display and then treat that as a black box and then trying to assess the quality of this black box that you have.

00:33:34.411 --> 00:33:36.218
Yeah, that makes sense.

00:33:36.218 --> 00:33:36.832
That makes sense.

00:33:36.832 --> 00:33:57.743
I was, I have not been thinking about this in such a way, but it is like testing a system in end-use conditions, something we need to do more in fire science, which you working at Frisbee, you're very, very well aware, because that's something that me and Grunder are battling for all the time.

00:33:57.743 --> 00:34:00.569
Good, good, very, very interesting.

00:34:00.569 --> 00:34:19.282
And the assessment is based on this image quality, the one that the test takes, and then apply some mathematical filters to extract some sort of outcomes of that, and then, based on that, it gives you a value, a number that indicates the performance of the device.

00:34:19.282 --> 00:34:20.143
Do I get it correct?

00:34:21.471 --> 00:34:31.418
So you more or less have this image and then you extract some breach of interest, apply some math we don't need to go into that and then in the end what you get is a single score.

00:34:31.418 --> 00:34:35.610
So this spatial response, or spatial- resolution.

00:34:37.295 --> 00:34:49.938
So what's the main challenge with this approach other than it's well, I guess being cumbersome and to some extent inconsistent is something that is very difficult for a testing method, so perhaps let's step over that.

00:34:49.938 --> 00:34:58.983
What is your view on how this can be improved and how we could improve the consistency of this assessment?

00:34:59.784 --> 00:35:02.829
Right, so that gets into the topic of the project models.

00:35:03.291 --> 00:35:12.485
So I guess the main question was what do other people, other industries, currently do in order to assess image quality?

00:35:13.130 --> 00:35:35.063
And then there's a huge field within image processing where you try to assess image quality, which is more or less the field that I then looked at and then saw okay, how can we extract some of this terminology and models and methods that they're currently using this terminology and models and methods that they're currently using and how can we apply that to thermal images and then try to use that for an image quality assessment?

00:35:35.063 --> 00:35:44.880
So I looked into multiple different approaches, but before that, I think we need to get some terminology, some common terminology.

00:35:44.880 --> 00:36:14.865
So when you have image quality assessments and you want to make sure that I want to have an assessment of this image and what I would then want to do is I want to either do something called a full reference image quality assessment, where I have an image that might have some distortions, let's say noise or blur or something and I want to compare that to an image of the same object, but with a very high quality.

00:36:14.865 --> 00:36:17.018
That's full reference image quality assessment.

00:36:17.369 --> 00:36:19.237
So I have an image that I then compare to something else.

00:36:20.090 --> 00:36:28.418
We can't use that for thermal imaging cameras, because if you take an image, you don't necessarily have a reference to then compare with it's more used.

00:36:28.418 --> 00:36:40.199
If you have, uh, for example, different algorithms to improve the resolution, then you want to perform some distortions and then see how can we improve the resolution of this poor image.

00:36:40.199 --> 00:36:41.101
Then you can compare.

00:36:42.170 --> 00:36:52.802
Unless you had a super high quality like order of magnitude quality better thermal camera that you could shoot the super cool reference picture, then you could probably do it.

00:36:52.802 --> 00:36:56.440
But I guess it could be challenging, right.

00:36:57.471 --> 00:37:17.632
So one of the other approaches is something called no reference image quality assessment, meaning that you take an image let's just say for now you take an image from your phone and we want to assess what is the quality of this compared to what a human might score it, so we can use some model that then tries to make a prediction of this.

00:37:17.632 --> 00:37:31.996
So that's the two main approaches, and I looked into both also because in the end of the project I had like sort of like a, a case study where I want to use these full reference metrics to then compare the no reference metrics.

00:37:31.996 --> 00:37:36.152
But let's skip ahead to the no reference image quality assessment.

00:37:36.152 --> 00:37:38.822
So I looked into three main approaches.

00:37:38.822 --> 00:37:42.355
So one of them is called natural scene statistics.

00:37:42.355 --> 00:37:47.675
Then you have machine learning-based models and then you have something called saliency-based models.

00:37:47.675 --> 00:37:50.974
So the first one, the natural scene statistics.

00:37:51.625 --> 00:37:58.655
Essentially what it does is it looks at statistical distributions of the image that you're trying to assess.

00:37:58.655 --> 00:38:08.715
So let's say you have an image, then you compute something called a wavelet coefficient or wavelet coefficients of the image and then you get some sort of distribution.

00:38:08.715 --> 00:38:18.713
It's not so important now the math that you do, but what people noticed was if you have different distortions, for example blur, you have a noise.

00:38:18.713 --> 00:38:23.351
It changes this distribution and that's essentially what the models are built on.

00:38:23.351 --> 00:38:26.925
Changes this distribution and that's essentially what the models are built on.

00:38:28.666 --> 00:38:31.900
So you can extract different features to then make a model that then computes some sort of image quality.

00:38:31.900 --> 00:38:40.858
And the whole backbone of this is that you have data sets where people have scored individual images with different distortion circles.

00:38:40.858 --> 00:38:45.059
It could be JPEG compression, you could have noise, you could have blur, you could have blur.

00:38:45.059 --> 00:38:49.786
You could have a very wide variety of distortions that you have on images.

00:38:49.786 --> 00:38:56.992
And then you have individual humans scoring the quality of this image and we can talk a bit about that later.

00:38:56.992 --> 00:39:07.590
But essentially you could have, let's just say, 100 people scoring different images with different levels of severity of the different distortions, and then you have an image dataset.

00:39:07.590 --> 00:39:09.771
In the end you can train models on this.

00:39:10.945 --> 00:39:14.110
But it's like an ambiguous scale, like everyone comes up with their own.

00:39:14.110 --> 00:39:17.289
For one person, the distortion could be nine, for one it could be three.

00:39:17.289 --> 00:39:21.713
Right, but you solve that by a large sample of you.

00:39:21.733 --> 00:39:36.001
You have a large sample and also what people normally do when they want to create one of these data sets is they have sort of like a calibration round just to sort of like base people in some scale that they can expect to be in.

00:39:36.706 --> 00:39:40.114
So this image is this bad and this image is this good?

00:39:40.114 --> 00:39:41.518
Approximately these scores.

00:39:41.518 --> 00:39:55.786
And then you sort of like calibrate the participants and then they go through a lot of images, they score that, and then in the end you'll have a data set where you have different distortions, different images and then different severity of the distortions.

00:39:55.786 --> 00:40:03.391
And then, once you have a lot of good data, you can then use that to train models on by train models, you mean machine learning.

00:40:04.186 --> 00:40:16.307
Also machine learning, but also some of the natural scene statistics models use support vector machines, for example, to then, I guess, map these features to the scores.

00:40:16.307 --> 00:40:18.088
But that leads us to the next one.

00:40:18.088 --> 00:40:25.251
So machine learning, and it's more or less, I guess the most common one would be convolutional neural networks.

00:40:25.251 --> 00:40:42.380
So you have a lot of different, I guess, machine learning architectures where people try to improve upon previous models and then once again try to see okay, if I have a score of this specific model, how do I compare it with the human score of this image?

00:40:42.380 --> 00:40:46.617
And then what's the correlation between, if I run this through the data set?

00:40:46.617 --> 00:40:54.318
So what you want to have is a high correlation between what the model predicts and also what a human would give this image.

00:40:54.318 --> 00:41:06.052
Because what this does is that you now directly have something where I can say a human would score the image, this specific value, but we also have a model that can do this.

00:41:06.052 --> 00:41:12.007
So now you actually have something that can predict the image quality of an image, of an arbitrary image.

00:41:12.007 --> 00:41:22.536
So that's the whole idea, and also the plan would be to do this also on thermal images, because if you have a very high correlation with this model and then human scores.

00:41:22.536 --> 00:41:32.811
What you can say is the model should score this pretty consistently and also, if you have this higher quality image, this should also be consistently higher than the lower image or the lower quality image.

00:41:32.811 --> 00:41:35.105
So that's the two first approaches.

00:41:35.346 --> 00:41:45.715
The last one is something called saliency, and this is a bit different because essentially what saliency models do is they predict where people look on images.

00:41:45.715 --> 00:41:57.016
So say, you have an image with some people and you have some faces Based on eye tracker data, there's large data sets where people have participated.

00:41:57.016 --> 00:42:04.434
Then if they look at a specific point of this image, then they get a point here, for example.

00:42:04.434 --> 00:42:12.855
So the idea with these salience models is to predict where people look on images, but once again removing the participants.

00:42:12.855 --> 00:42:34.650
So train a model and then you can say, okay, based on this model, trained on this data, with eye-trigger data, we predict that a person will look at this point in the image and the way people use this for image quality assessment, because it's not necessarily immediately trivial to see why that would blend into image quality assessment.

00:42:35.331 --> 00:42:44.998
But what people saw was that if you have this saliency model and it creates a saliency map, which is essentially just probability of where people look on the image.

00:42:44.998 --> 00:42:55.451
Then they can use the saliency map with another image quality assessment model to then boost the performance of the model by a few points.

00:42:55.451 --> 00:43:04.534
So then in the end what you get is a higher correlation between the model that you trained with the saliency map and then the human score from the data set.

00:43:04.534 --> 00:43:09.836
So those are some of the models that I looked into.

00:43:09.836 --> 00:43:11.431
So these three main approaches.

00:43:12.704 --> 00:43:21.853
I have so many more questions about thermal imaging, but let's close the report because I think we're narrowing to the conclusions and giving those three techniques.

00:43:21.853 --> 00:43:28.894
What were your concluding recommendations to improving the quality of the assessment of those cameras?

00:43:29.677 --> 00:43:29.878
Right.

00:43:29.878 --> 00:43:37.373
So what I saw when I looked into the different models and also the datasets was that a lot of work in the visible spectrum.

00:43:37.373 --> 00:43:47.393
But I've only saw maybe one study, I think, that created its own dataset and then trained a model on that dataset to then predict the thermal image quality.

00:43:47.393 --> 00:44:02.597
So there is some work on it and it can be done even with some of the earlier models the natural scene statistics but at the moment there's a gap in the research because we don't actually have a thermal image quality assessment dataset.

00:44:02.597 --> 00:44:06.018
We don't have a dataset with also it could be for firefighting, but also just in general we don't actually have a thermal image quality assessment data set.

00:44:06.018 --> 00:44:13.498
We don't have a data set with also it could be for firefighting, but also just in general we don't have a data set with thermal images where people have scored the quality of this specific image.

00:44:14.105 --> 00:44:15.349
What would constitute a data set?

00:44:15.349 --> 00:44:16.373
What's a data set?

00:44:16.373 --> 00:44:16.885
So?

00:44:16.985 --> 00:44:23.425
imagine you put a bunch of people into a room and you have some images in the thermal spectrum.

00:44:23.425 --> 00:44:25.472
So you have a lot of different thermal images.

00:44:25.472 --> 00:44:28.152
Then you degrade that to some degree.

00:44:28.152 --> 00:44:39.934
So let's say we add blur, we add noise, we add some JPEG compression, we ask people to give a specific score of all of these images and then in the end we aggregate that.

00:44:39.934 --> 00:44:53.960
So you have one big data set, so all of the different images named and then also the corresponding scores from the human that would constitute the data set and what metrics make it a good data set versus a bad data set?

00:44:54.460 --> 00:45:02.282
I mean you need to have a certain amount of people, but also it's a difficult question because a lot of the work was in the visible spectrum.

00:45:02.282 --> 00:45:09.943
But you need a certain amount of representative distortions that you expect to see on thermal images in general.

00:45:09.943 --> 00:45:12.612
You also need it, especially if you go.

00:45:12.612 --> 00:45:20.733
If you want to apply this to the fire service and it should be relevant for firefighters then you also need to have some conditions.

00:45:20.733 --> 00:45:26.132
That's representative of the scenarios that a firefighter will expect.

00:45:26.132 --> 00:45:28.588
So it's in the visible spectrum.

00:45:28.668 --> 00:45:31.695
People were sitting in front of monitors.

00:45:31.695 --> 00:45:33.099
Those are calibrated.

00:45:33.099 --> 00:45:39.525
You had an evenly illuminated environment, but that environment is not necessarily the same.

00:45:39.525 --> 00:45:53.681
If you want to have a dataset that's also fitting for firefighters and the thermal image quality in the fire service, perhaps you need to use monitors that are representative of the thermal imaging cameras.

00:45:53.681 --> 00:45:56.875
It may be three, three and a half inches.

00:45:56.875 --> 00:46:10.456
So there's a lot of different things and also the distortions that you might expect is also different for infrared images, right, so you might have some radiation reflected off whatever surfaces and so on.

00:46:10.456 --> 00:46:25.137
But in the report we try to give some recommendations on how to create a dataset and also what we should think about if it should be relevant for thermal images and also some of the challenges that you might run into in the end.

00:46:25.625 --> 00:46:26.427
A follow-up question.

00:46:26.427 --> 00:46:26.829
Is that?

00:46:26.829 --> 00:46:46.376
Okay, you're talking here a lot about image quality in relationship to fire fighting, which is absolutely understandable if we're talking NFPA 1801 thermal cameras, but that's not an entire use of thermal cameras in fire safety and I would argue there are, for example, detecting a fire.

00:46:46.376 --> 00:46:55.353
I think that is a very interesting use of thermal cameras which is currently not your mainstream way of detecting fires.

00:46:55.353 --> 00:46:57.713
I had an episode on detection a few episodes ago.

00:46:57.713 --> 00:47:19.706
I had a very interesting episode on waste fires with Ryan Fogelman, who is the globe's number one refurbisher of FLIR cameras, and the reason for that is Ryan is using them to detect fires in his devices in the fire over towers, so they are using thermal cameras to detect fires in the waste facilities.

00:47:19.706 --> 00:47:48.862
If you've missed that episode, I recommend going there because we talk a lot about practical use of this technology in that episode and I think you know having the ability to stand in a standardized way, in a controlled manner, define the minimum characteristics of a thermal camera is also one of the barricades on the road towards using them as a reliable detection device.

00:47:49.385 --> 00:47:55.338
And, what's interesting, you said that the resolution of your cameras in 320 to 140,.

00:47:55.338 --> 00:48:09.500
You've classified this as low, and I agree if I want to look at an image, but if I'm trying to cover a field of view that spans over a building or something and I just want to know is there a fire in the room or is there not a fire in the room?

00:48:09.500 --> 00:48:14.956
Perhaps a simple information any pixel of them is overheated over my fire temperature.

00:48:14.956 --> 00:48:19.876
Maybe I can go with a camera of a resolution of, I don't know, 50 by 50, 20 by 20.

00:48:19.876 --> 00:48:21.351
I don't know how low you can go.

00:48:21.351 --> 00:48:29.956
I know people who go extremely low with the resolution and yet are capable of getting some sort of information through their system.

00:48:29.956 --> 00:48:56.166
So do you think those metrics and those testing approaches perhaps minus the people assessing the data, because that's very you know, I observe the image type of assessment, but in terms of gathering data resolution, I think we'll also need to come up with tests like that to check the robustness of the future infrared fire detectors really Right?

00:48:56.206 --> 00:48:56.969
And I completely agree.

00:48:56.969 --> 00:49:04.302
I mean you don't need a full HD image in the infrared spectrum in order to do a lot of the tasks that we use.

00:49:04.302 --> 00:49:11.938
Thermal imaging cameras for, absolutely, as you say, go lower, also depending on what you need to do with the camera.

00:49:11.938 --> 00:49:17.876
So the current resolution is more than enough, I believe, to go into burning buildings.

00:49:17.876 --> 00:49:40.793
I think that the question is not so much should we improve the quality of the cameras for this specific purpose, but more do we have a way that's actually robust, predictable, consistent to test the quality, because you don't want to have a mishap, for example, where you have to fail a camera because of the tests, because of the inherent inconsistencies in the test.

00:49:40.793 --> 00:49:43.474
You just want to have a robust method.

00:49:44.264 --> 00:49:51.132
I think what this report also opens up to is perhaps more interesting question is what else could we use such a dataset for?

00:49:51.385 --> 00:50:15.525
Because there was a recently published report by NIST that also talks about AI and machine learning in the FHIR service, and perhaps it's not so much the quality that should be in the focus from this report, but maybe the fact that we don't have a dataset of thermal images of firefighting scenarios, because if you have that, you could apply that to a lot of different things.

00:50:15.525 --> 00:50:19.215
You could apply machine learning directly on the thermal imaging camera.

00:50:19.215 --> 00:50:21.010
You could try to detect humans.

00:50:21.010 --> 00:50:23.360
You could improve split-second decisions.

00:50:23.360 --> 00:50:25.264
You could try to detect humans, you could improve split-second decisions.

00:50:25.264 --> 00:50:46.976
And if you have a data set, that will not only open up for this image quality assessment, which I believe is also valuable in itself, but perhaps more for research purposes and this consistency issues, but opening up the door to a lot of difference and other opportunities applying machine learning to who knows what right.

00:50:46.976 --> 00:50:50.231

214 - Thermal Imagers with Martin Veit

Listen On

Recent Episodes

Support On