FLIR New Coronavirus Prioritized Temperature Screening Camera Examined

By Ethan Ace, Published Apr 03, 2020, 10:07am EDT (Info+)

FLIR has announced a new series of thermal cameras "prioritized for entities working to mitigate the spread of COVID-19 virus", the A400/A700 series.

IPVM Image

In this post we examine this new camera line, how it compares to FLIR's other options and competitors, and outlook for FLIR.

A400/A700 ********

******* (*******) *** **** (*******)*** ******* ***-***** ******* **** ******* lens ******* **** *° ** **°. They **** ** **-***** ******* **** many ** ****'* *******, **** ** the* ********* ************, *** ******* *** ******** *** Ethernet ** **-**.

Accuracy ****** / ** ***** **** ********

** ********* ****, **** ****** ******** of ±*.*° * (~±*.*° *). **** accuracy **** ** ******* ** ***** other * *** * ****** ****** as **** ** **** *********** ********* systems ** **** ******** **** ***************. *******, ****** ****** ******** ±*.*°* or *****, **** **** *** ******* a ***** **** **********, *********** *** regular *********** *** ****** ******** *** proper ***** **** *** *** ******** installation *** ***** ****.

**** **** ****'* ****/**** **** ****** list ******** ** ±*° *, *** this ** *** ******* ******** **** of *** ****** ****** *** ****** temperature ***** ** ** ******* ** measuring, -**° * ** *,***°* (-*° to *,***° *), *** ******** ** screening ****. ********* **** *********** ** not ********* ** ***** **** ******.

Stop *** ***** ********* / ** **-***-**** ***********

***** **** *******, ******** *** ******** one ** * ****, ** * distance ** *-** (~*.*-*.*') ***** ******* lens *******. ******** **** ****** ******* and ****** ***** *** *********, ***** takes *-* *******, ***** ** *** video *****.

********** ********* ******** ********, ** *** camera ** ****** **** ** ******* small ******** ** ****** ******* **** specific ****** ** *** ****, *** eases ******** ******** ******** ** ****** which ***** **-***-**** ***********, ***** *** subject *** ** ****** *** ** line *********** *** ********* *********, ******* of ******** **** ** * ****** crowd.

*******, ** **** ********** ************ **** as ******** ** ****** ***** ******* areas, **** *********** *** ***** ***********, requiring ********** ********* ******* ** ******** throughput.

Inner ******* ** *** *** ********

**** ******** ********** ********* *** ******** glasses ** ** ******* ******* ******* temperature ** ******** ***** ******** ******* ** *** ***, * ***** ****** **** *** tear *****, ******* ** ********* *** skin ** *** ****/******** **** **** competitive *******.

IPVM Image

*** ***** ******* **** ** **** more ******* ********** ** ****** **** temperature **** ***** ******* ** *** face ******* ** ** **** *** ophthalmic ******, ***** ********* ***** ****, and ******* ** ** **** *********** to ************* ******* **** ** ****, cold, ** ******** ***** *** ***** inaccurate ************ ** *** ****.

IPVM Image

*** ******** ** **** ******* ***** / ******* *** ***** ******* **** thermal *******, ********* ******* ** ** removed ** ******* **** *********, ***** is ************ ** ****'* ****** ********* demo *****:

Adaptive ******** ************

**** **** ****** ** ******* ******** by ********* ******** ************ **** ****. A ******** ** *********** ***** *****-******* subjects (**********) ** *****, *** *** ****** automatically ******* *********** ******* ** ********* if ** ******* ******** ******** ********* or **********.

**** ****** **** ** *** **-*********** which ****** ********* ** ********* ********** the *** *** ** ******** ******* conditions *** ** *******, **** **********, crowd ****, ***. **** *********** ***** to ******* **** ********* **********, ** well, *** ** ** *** * common *******.

FDA ***(*) *********

**** *** ******** *** ***(*) ********* for ***** ******** **** *********** *******, allowing **** ** ** **** *** adjunct ********* (********* * ****** *********** with ** ******** ***********). **** ** uncommon ***** *********** *******, ****************** ***** **** ***** ************* ********* ******** body *********** ********* ***** **** ******** clearance (*** *** ***'* ******* ** ***(*) cleared ******).

********* ****** ******* *** ********* ** risky, ** *********** ******* ***** (*** historically ****) ******** ** ******* ******* and *****. *** **** *********** ** FDA **********, ***:*** ***** ******** ** '***********' ******* Fever ********* *******.

*******

**** *** **** ****** ******* ****** from ~$*,*** *** ***** ********** **** models ** ~$**,*** *** *** ******, also ********* ** ****. **** ** priced ******* ** ** ********** ***** than **** *********** ******* ***** ******* camera *** ***** **** **********, ***** sell *** $**,***-**,***. *******, **** ** 4-5x *** **** ** ******* ******* such ****** ****($*,*** ****).

Prioritizing *********** ********

*** **** *** **** *** ********* for ******** ***. *******, **** ** currently ************ **** **** *** ***** working ** ******** *** ****** ** COVID-19. ***** ******** *** ***** *** cameras, *** **** *** ** *********** until ******* *******/****** ****** ****** *** filled.

*******

****'* **** *** **** *** *********** very **********, **** *** ********* **** competitors ** *** ****, *** ******* similar ** ********* *******/******* **** *********** with **** **** ******* ** *********** measurement, **** ** *****, *********, *** Sunell. ************, ***** *** ** ********** screening ** *** ******* ** *** eye ** ************* ******** **** ***** systems ******** ** *** **** *********** of *** ****.

*******, ***** ******* ** ***** **** most ***********, ***** ******* *** * quick, *********** ******** ** * ******* which ***** *** ******* ** * matter ** * *** ****** *** find $*,***+ *** **** *** **** lower **** ************ **** ** ****.

Vote / ****

UPDATE ********* *********

**** **** ***** ******* ** ******* a ********* ****** **** ** *** generally ********* **, ******:

*** *** ****/**** ***** ******* **** EST ************* **** ** ****** ********* mode, *** ** **** *************, ** not ******* * **. *** *** A4xx/A7xx ***** ******* ** ******* *** use ** * ** ********* ** users **** ** *** **** ******. To ** **, *** ****** **** on *** ** ********* ** *** camera ******* ********* *** ***** *** cursor ** *** ** *********. **** your ********* *********** **** **** *** that *********.

***** *** **** ********* **** ****** EST ****** *************. ** ******* *** screening **** *** **** ******** **********, two ** ***** **’** ****** (*****), but *** ***** **** **** ** use * **, ** ******* **** configuration ** ****:

  • ******** ** +/-*.**, ***** ***** *** FDA *** *** ****
  • *********** *** ********** ** **** ************ throughout *** ***, ** ***** *** last *** ******* ******** ** *** reference
  • **** *** ******* ******* ***** ** hardware ** ******** *** ********* ** the ******.

Comments (38)

Are these traditional IP cameras? E.g. can we get an RTSP stream?

It appears so, but is there some VMS or other device that provides the feedback/alerting on fevers?

Agree
Disagree
Informative
Unhelpful
Funny

Yes. They do have an RTSP stream. Also, while FLIR is not claiming ONVIF conformance, some distributors I've seen are mentioning it. I've asked FLIR for clarity on that.

Note that they have M12 outputs on the rear, not a typical Ethernet jack, but they also offer an M12/RJ45 adapter cable.

Agree
Disagree
Informative: 15
Unhelpful
Funny

Thanks for the heads up on this, Ethan. Great article.

Agree: 4
Disagree
Informative
Unhelpful
Funny

Great article and assessment, thanks! I reached out to our FLIR rep yesterday and was surprised that Security Integrators can not buy this product. It is not classified as a security product line item. Since Integrators are focused on security and safety - - this seems odd. Several of our clients are requesting this level of additional screening (on top of their access control) to just have one more layer of security.

Such shame that our hospitals and critical infrastructure clients can't purchase through their trusted integrator.

Agree: 1
Disagree
Informative: 9
Unhelpful
Funny

This is just one of many reasons why FLIR's Security group has had no real growth over the past several years. Internally, FLIR is split up into a number of divisions. This camera is part of the Instruments group. Security is in the Commercial group. The Instruments group more or less believes any radiometric product is an "instrument" and not a "security device". They have little to no incentive to potentially lose business to their peers in a different division of the company.

You will also find the Government group doing a product+installation bid for a large project like an airport, effectively competing with the Commercial group offering the same products via the security channel to the same customer.

FLIR Security is all but dead at this point. If you are interested in this product, find a contact in the Instruments group and they will help you purchase it through their channels.

Agree: 4
Disagree: 6
Informative: 10
Unhelpful
Funny

For the disagree's, I would love to hear your counterpoint.

Agree: 4
Disagree
Informative
Unhelpful
Funny

We are a FLIR distributor for these cameras and can work with you, if you like.

Leave me you contact info here and ask for me in your request.

Contact us - MoviTHERM

Agree
Disagree
Informative: 1
Unhelpful
Funny

Markus,

I'm interested in learning more about the two FLIR devices mentioned in this article.

Thanks,

Drew

andrew.schrader@us.panasonic.com

Agree
Disagree
Informative
Unhelpful
Funny: 2

You sure it's 0.3°C? In link provided it says it's +-2°C

Also in materials you posted before about temperature measurment there was a link from FLIR that for +-1°C-1% you need blackbody:
Infrared Camera Accuracy and Uncertainty in Plain Language | FLIR Systems

Agree
Disagree
Informative: 1
Unhelpful
Funny

FLIR clarified this. The specs are a bit confusing. The ±2° spec is the generic accuracy of the camera, across the the entire temperature range it is capable of measuring, -20° C to 2,000°C (-4° to 3,632° F).

In elevated temperature screening mode, accuracy increases to ±0.3°C, as the camera is used in a stable ambient environment, only humans are measured (instead of other objects in a wide FOV), and their reference samples are updated according to the population being screened (adaptive baselines discussed in the post). This is not reflected in the spec sheet, but FLIR discussed this with us.

We've updated the body of the post for clarity.

Agree
Disagree
Informative: 6
Unhelpful
Funny

How could FLIR be as accurate as Dahua? This does not make sense, right?

Agree: 1
Disagree
Informative
Unhelpful
Funny: 9

More like "how it can be as accurate as medical cameras that require black body for thermography". But if they managed to do this using some other reference and software - thats great! It means that black body is not needed and this solution will be cheaper, without need of re-calibrating black body every 1-2 years. And that would mean that more people can afford such solution. If it's true, I'm all in!

Agree: 2
Disagree: 2
Informative
Unhelpful: 1
Funny

So do you have any link to this statement? Like on their webpage? I would need something official you know ;)
Also, this paper Infrared Camera Accuracy and Uncertainty in Plain Language | FLIR Systems
In this study they show that even with camera looking on blackbody alone, re-calibrating all the time, with stable moisture, temperature and so on, they can achieve around 0.3-0.6°C.
So now, if I get this right, they are using "people" as this "blackbody reference" to re-calibrate camera all the time? What will happen if someone closes shop for weekend, what will happen with this calibration during this time? It will re-calibrate on first person that enters the shop? So it means that at least first people entering the shop should be checked with some thermometers to make sure that their temperature is stable. Is there any more info, manual or something from FLIR? I cant find anything more on their website.

Agree
Disagree
Informative
Unhelpful
Funny

We've asked FLIR for documentation. They are not using people as black body reference, exactly. They are inspecting a rolling average to see if there is a temperature trend up or down which might indicate an ambient environment change to determine if alert temperatures should be raised or lowered. Each subject's temperature measurement is absolute, not relative to others.

Also, if you close a shop and re-open Monday, yes, you would need to re-establish baseline. But all these systems should be rechecked/recalibrated, at least daily. Dahua's own documentation for their body temperature system mentions this.

Some companies we have talked to who have been in the temperature measurement field for years say that these checks should be multiple times per day, even when using a black body (which should also be checked for accuracy and recalibrated periodically). These are critical measurements, not set it and forget it camera systems.

Agree
Disagree
Informative: 2
Unhelpful
Funny

Thank you for this article. Do you know if the advanced is required for EBT?

Agree
Disagree
Informative
Unhelpful
Funny

I asked FLIR to clarify that along with some other questions above. They haven't yet responded. I'll post back here as soon as we get response.

Agree
Disagree
Informative: 1
Unhelpful
Funny

Hi Shaun!

We are a distributor and systems integrator for these cameras.
There is currently no specific EBT version of the FLIR A400/A700. The smart sensor configuration (Advanced) of the A400/700 offers a black body compensation function. In other words, if you are placing a region of interest on a reference black body, all measurements will be corrected against that source.

You could use a region of interest and place it around the eye and then query for the max. temperature, which will be the canthus, as long as you don't include the forehead.

Here more on that camera:

FLIR A400 A700 Smart Sensor Thermal Camera - MoviTHERM

Agree
Disagree
Informative: 2
Unhelpful
Funny

Funny how the Dahua thermal camera gets slammed for just about everything, and here everyone is intrigued by Flir's offering. Not defending Dahua or picking on Flir, but we all know this is pretty much useless in stopping the spread of the virus.

Agree: 4
Disagree
Informative
Unhelpful
Funny: 1

Brian, I think that's a fair criticism. A key difference though is expectations, FLIR has been very conservative.

FLIR says people need to stop, they need to take off their glasses, they need to look at the camera, etc.

Watch their video:

Dahua shows images of masses of people getting off a train, continues to show people walking quickly past the camera, etc.

To the extent that these solutions work, I have a lot more confidence in ones that responsibly market it than ones who do not. Yes/no?

Agree: 9
Disagree
Informative
Unhelpful
Funny

While yes this may be more responsible marketing, it does not accurately detect those infected the virus. As we know now the asymptomatic people are able to spread the virus just as easily as those with symptoms. So while this may catch symptomatic people it's not really helpful in stopping the spread.

Agree: 4
Disagree: 1
Informative
Unhelpful
Funny

I disagree with the statement: "So while this may catch symptomatic people it's not really helpful in stopping the spread."

It is true that no camera (or thermometer for that matter) will be able to detect an asymptomatic person. And no thermal camera has actually been approved by the FDA to even "detect a fever". They are merely capable of detecting "variations in skin surface temperature".

However, there have been studies done on the effectiveness of screening (with all things considered). The statistical models show that screening is more effective than not screening. Even, if you only catch symptomatic folks.

They even publish data a simulator, if you would like to change your assumptions on things and see how this affects the spread.

Estimated effectiveness of symptom and risk screening to prevent the spread of COVID-19 | eLife

Agree: 2
Disagree
Informative: 3
Unhelpful
Funny

The problem is this doesn't confirm if people have a fever or not. It only identifies people with elevated surface temperatures. So while someone may appear hot on the surface, it doesn't necessarily mean they have a fever. If this is true, could the opposite be true? A person with a fever might not show elevated surface temperatures. Again, this is merely one symptom of having the fever and if you're looking to do risk screening, using a camera to identify people just isn't practical. Especially at the price point they are asking for these cameras. To be effective, you'll need to screen the people by asking questions, and if you've got them stopped to ask questions why not just use a $75 laser thermometer instead of a $10,000 camera? You would have to force people in to one line to stand in front of a $10,000 camera for a few seconds or you could have over 130 lines each with laser thermometers for the same equipment cost.

Let's say you have something like this setup at the entrance to a nursing home and an asymptomatic person with the virus passes through with no fever. This person infects the entire nursing home. You think they are going to be happy with their purchase?

Dahua's marketing claims to catch people on the move in crowds may be false advertising, but that is the type of system that would make the most sense. IF you could accurately scan crowds of people and pick out those with elevated surface temps you might be able to divert them in to a secondary screening line for further checking. But again this is only one symptom.

Agree
Disagree
Informative
Unhelpful
Funny

You need to calculate also labor cost. 130 lines = at least 130 people (that are at risk, because they are close to those screened people), if it works 24h, this means 130x3 plus some coffee/lunch break.

I'm still reading those studies posted by #8 but it looks like screening is not so bad idea.

As for Dahua or Hikvision marketing - for me at least some of those videos look like testing videos. I hope that ipvm will test some of those sets and will verify them. Especially those without black-body radiator.

Agree
Disagree
Informative
Unhelpful
Funny

Of course labor cost would be a factor. I was simply pointing out that the cost of a single camera makes it a bit excessive for practical use for what little it accomplishes.

Agree
Disagree
Informative
Unhelpful
Funny

A disease of asymptomatic spreaders (50%?) without masks and deniers. A bad combination.

Agree
Disagree
Informative
Unhelpful
Funny

The article says you examined the Dahua and Hik solutions? Have you actually bought and tested these or is this a mis-leading statement?

The intent of all these solutions i believe to be good. Looking for elevated temperatures that put people and places at risk. Not only for today but moving forward having solutions that can alarm on elevated temps from a distance without creating lines to warn of those that may pose a risk to workers, customers and facilities. It is not about alarming on Covid but alarming on those that have elevated temps.

It seems this solution has no face rec. and requires an operator to make sure the persons head fills the FOV.

So very interesting that there is no trashy headline on FLIR.

Very clear how a bias effects headlines and the review. This is actually a review but it seems from marketing material vs testing.

Agree
Disagree: 2
Informative
Unhelpful
Funny

Very clear how a bias effects

You are a Dahua rep who has made 2 total comments, both defending Dahua, so let's keep that in mind when accusations of bias are being thrown out.

We have not yet tested anything, FLIR included, because virtually everything is sold out and backordered.

can alarm on elevated temps from a distance without creating lines to warn of those that may pose a risk [emphasis added]

Not creating lines is a great thing to strive for but it's obviously going to impact accuracy. People walking and looking in different directions (to the left or right or down looking at a phone) or with a ballcap on or hair covering their forehead are obviously going to reduce the system's performance. The reason why we did not criticize FLIR is that they admit that.

And FLIR is aiming to measure on the inner canthus which is clearly more accurate than the forehead. Does that matter to Dahua reps?

Agree: 4
Disagree
Informative: 1
Unhelpful
Funny: 2

Don't be defensive. I am pointing out a clear bias.

Nothing is tested but very interesting how 2 articles have a a very different tone. I'm just looking for no trashy headlines, articles that point out the positive and drawbacks of each solution from whomever.

Maybe i expect too much, like an industry consumer reports.

I am fine with you stating i am a Dahua rep, no problem but should serve as a warning to all that post undisclosed is not so undisclosed. Might be concerning for some.

Agree: 2
Disagree: 1
Informative
Unhelpful
Funny

I am fine with you stating i am a Dahua rep, no problem but should serve as a warning to all that post undisclosed is not so undisclosed. Might be concerning for some.

Undisclosed has always been this way. We don't disclose names but affiliations are important to understand, in your words, bias. For example, if a FLIR rep undisclosed attacked Dahua, we would call that out too.

very interesting how 2 articles have a a very different tone.

It's pretty simple. FLIR has a very conservative and technically grounded approach. Dahua leads its marketing with a faked image and sweeping claims.

And before you go to the USA vs China thing, we fought a USA CEO recently over their faked coronavirus camera marketing.

Agree: 2
Disagree: 2
Informative
Unhelpful
Funny

I strongly disagree with the bias effect on this article. As John stated, he's responding to conservative marketing vs. ridiculous claims or false advertising as others have done. If Flir announced they have the perfect solution in this and didn't give the clear examples, I think the headline would have been extremely negative and maybe even harsher than Dahua and Hik because it was Flir.

Agree: 4
Disagree
Informative
Unhelpful
Funny

The intent of all these solutions i believe to be good.

I disagree. The intent behind many of them appears to be rushed products to try and capitalize on fear.

If this body temperature solution did not exist 3 months ago, then (IMO), it is a rushed product that is likely not going to be accurate or reliable over time.

If it is a recent product, post detailed data on how/why it was developed and evaluated specifically for this application. Explain why the microbolometer used was chosen, how it is calibrated at the factory, guidelines for setup and managing emissivity changes in various environments, etc.

If it is not a recent product and it has been around for a while, post some case studies or white papers that clearly show previous success and applications.

Agree: 8
Disagree
Informative: 1
Unhelpful
Funny

Any update on FLIR offering? Any version that connect ONVIF or otherwise with VMSs such as Genetec and Avigilon, Milestone?

Agree
Disagree
Informative
Unhelpful
Funny

FLIR tells us that the A400/A700 cameras will be ONVIF conformant, but they are not yet on ONVIF's official listing.

Agree
Disagree
Informative: 1
Unhelpful
Funny

Could someone from FLIR please reach out to me on this product? Thanks!

Agree
Disagree
Informative
Unhelpful
Funny

You have just identified the major issue with the product ;)

Agree: 1
Disagree
Informative
Unhelpful
Funny: 2

Has anyone understood whether the FLIR A400 / A700 series cameras are or are not "bi-spectrum"?The datasheet shows the "non-thermal" component as optional ... but if it is present, shouldn't there be separate lenses on the front of the camera?I have never seen any photos of the A400 series with two lenses ....

IPVM Image

Agree
Disagree
Informative
Unhelpful
Funny

The visible camera is at the bottom of the front of the camera, actually. It's the small dot at below the thermal lens.

IPVM Image

Agree
Disagree
Informative
Unhelpful
Funny

Integrators be Aware that FLIR will sell this solution DIRECT to your customers!

I Introduced the FLIR A400 solution to a large national account of mine and that customer did reach out directly to FLIR and FLIR quoted the large amount of units and they are selling direct!

Not Cool!

Agree: 1
Disagree
Informative: 2
Unhelpful
Funny
Login to read this IPVM report.
Why do I need to log in?
IPVM conducts reporting, tutorials and software funded by subscriber's payments enabling us to offer the most independent, accurate and in-depth information.
Loading Related Reports