Axis Thermal Camera / Grekkom Human Temperature Control Tested

By Ethan Ace and Derek Ward, Published Jun 10, 2020, 01:58pm EDT (Research)

***** **** *** * *********** ********* camera (********-*), **** *** **** ********* ** market ** *** ***********. *******, **** partner ******* *** **** ********* *** '***** *********** *******' ****** **** *****-*. *** *** **** does **** ****?

IPVM Image

** ****** *** ******* *** *** on ** **** *****-*, ********* *** following:

  • *** **** *** ******* *** *** detect ** **** ******** **** ***********?
  • *** **** *** ****** ****** ****, masks, *** *******?
  • *** ** ******* *********** ** *** move?
  • *** ***** **** ** **** ** be **********, *** *** **** **** calibration ****?
  • *** **** ******* ******** ****** ******** temperature?
  • *** **** ** ****** ******** ******** in *** *****?
  • *** **** *********** ******* ** *******'* marketing ******?

Executive *******

** *** *******, *** ******* *** application ******** ************ ****** ±*°* ** an ** ***********, ******* ** ***** claimed ±*.*°* / ±*.*°* ***** *** lower ******** **** **** ***** ********* systems ** **** ******.

************, ******* **** **** **** **** a "******** *******" ** ************ ** key ****** ** *** **** (********, eyes, *****, ***.) ***** ******** ** missed ********* ** ******** ************ ** our ***** **** ******** **** ************ warmed ** ***/***-***** ***** ************ (~***.*-***°*) as ****** ***** ** *** **** brought **** ******* ***** *********.

*** *** *** ******** ************* ******* times *** *** ** *** ******* as ******* ************ **** *** ****, a ~**-** ****** *******. *******, ***** is ** ********** ** ***** **** the *** ** *** ** *********** other **** *********** ******** ********, ***** is ****** ** ******** ****** ********** and/or ***** *********.

**UPDATE */**/**** Grekkom has contacted us asking us to re-test their HTC application as they say the version is out of date and the results are invalid. Per their response below:

** ****,

***** ** *** * **** ** thank *** *** ****** *** **** to ** ** **-***** ****** ** our ********. ** *** ***** ** see **** **** ** **** ******* results **** *** ** ************ ** expected, ******* * ******* **** **** is *** ** * *********** ** unfortunate *******.

** *** **** *** ** *** present ********* ***** ** * **** demand ** ***** ***** ** *********, which ***** **** *** ********* ** a ****** *** ********* *** *** review ******* *** **** ********** ******* through *** *****. ******* ** ****, not **** ** ******'* **** *** time ** ***** *** ******* *** correct *** ** *** ********, *** also ** ******** *** **** ** outdated ******* ** **.

** *** ****** ****** ******* ** have **** ** *** ****** ********* in *** ****** *****, **** ** the **** ** ****** *** ************* when *** ******** ** ***. **** there ** * ****** ** ***** that ********** ** * ****** ********* and ********* *********. ** **** ******** a ****** ** ******* ******* ** customize *** *** *** **** ******** the *******.

******* ****** **** *********** ** **** of *** *** ** ******** ******* of *** ****** ** *** **** of **** ***** ***** ** ** some ********* ***** *** ******** ******** we ********* **** ** ******** ************ detections *** ******* **********, ***** ******** are **** ********* *** *** ****** based ********, *** **** ***** *** only ****** *** ****** ** * time *** ** ****** ********* *** indoor *********.

*** ****** **** ******* ** *** most ** *** ******* ***** *** precision, ** ********* ********* ******* **** thousands ** ****** *** ** ******* countries **** ********* **********, ******* ***... Before ********* *** ******** **** **** Q2901-e, ** ********* * **** ******* to ****** ***** **** ****** ********* and ** ***** *** **** **** level ** ******** **** ********* **********. At **** ***** *** ******** ** being **** ** * **** ****** of ******* *** *** **** ***** happy **** *** ******* **** **** learn *** ** **** **** **. I'm **** **** *** ******** ***** anomalous ****** *** * ****** ** it *** *** ** * **************** with *** *********** *********.

*** ***** ******* ** *** ********* if *** ***** ** ****** ** extend **** ******* *** *** *** conclusions **** *** **** ****** ******* and **** *** **** ** *** of *** *********** ** ***** *** through ** *********** ******* *** ******* your *********.

**** *******

*******, ****** *******, ** ********* ******* with ********* *** ******** ***** *** our ********, ** **** **** ***** aware *** ******** ** ******** **********. Additionally, * ******* ******** ******** ********* and ************ *** *********** *** **, so *** ******* ** ****** ************* were ** ***** ** *** ***.

******** */**/**: ******* *** ******* *.* Tested

**** *** ****** ******* *.* ** the ******* *** ***. *** **** temperature *********** ******* ** *** ****, with ******* ***** * ******** ******* and *********** ******* ** ** **** when ***** ****** * *********** ********** in ******** ***********. ****** *** ***** issues **'** ******* ***** *** ***** present, **** ** *********** ** *****/****, influence ** ******** ************ ** ***** outside, *** ****/*****/**** ******** ******** **** temperatures.

** **** ******* **** ** * subject **** * **.*°* ***********, *** HTC *********** *** **** ** **** within ±*.*°* ** **** ******* ****, which ** **** ****** **** **** we *** **** **********.

*******, *** ***** ****** ***** ******, and *** **** ***** ***** ******* have **** ** *** *** ******** controls, ******** *** ****** **** ** an ***** *******, ******* **** ****/******, and * ****** *** *******/****** *** temperature *********** *** ****.

Axis *****-* **** *********** ***********

***** *** ******* *********** **** ***** the **** *****-* **** *** *******, without ***-** ****.

  • ******* ************/**********:*****, **** ******* ************ *** ********, e.g., **°*, **°*, ***°*, **** ** option *** *******, ****** ************ **** less *******. ***** ********** **** **** use ********, ** ***** **** *** alarm ****** ** ***° ** ***°, for *******, ******* ** ***.*°*.
  • ** *********** ******:****** ******* ***** ******** * *********** measurement **** ******** *** "******" ***********, the **** ****** ***** **** **** provide ****-*********** ******.
  • ** *** ***** *********:*** **** *****-* **** *** ************* display *** ******* ***** ** *** camera's ********* ****, **** ********* ****** when ********* *******. ***** **** ******** click ** ******** **** ******** (*** below).

************, ************ **** ********** **** ***** only *** ******. *** ******* ***** measured ** ***°* *** ~**.*°* ** an ** ***********, **** *°* *****.

IPVM Image

Grekkom ******** ******* ********

*** *********** ***********, *** *** *** algorithm **** * "******** *******" ** exposed ****, ********* ** ******* **:

*** ********* ******** * ******** ******* of *** ******* **** ***********. ** weighted * **** **** ***** **** the ****, *** ***** ** *** forehead **** ******* **** * **** significant ****** ** *** ****** *****.

** *** *******, **** ********* *** to **** ***** **** ***** **** a ******-**-******** ***** ***** *********** **** through *** ****** ******* ********** **. Shown *****, *** ******* ** ******** at **.*°* **** *** **** **** is *******, *** ******** ** ***.*°* using ** ** ***********. ***** ******** cooler ***** ** *** **** **** a ****, ******* ******** *** ** 101.6°F, ********** ** *****.

IPVM Image

*******, ***** *********** *** ***************** **** ***********. ** *** ******* below, *** ******* ** ******** ** a "******" **.*°* **** ******** *** scene. *******, ******** *** ***, *** temperature ********* *** * **** *********** alarm ** *********.

No **** ********* / *** ******* ********

*** ******* *** **** *** *** face ********* ** ********* *********** ******. In *** *****, **** *** ** other ******* ***** ******** ** **** entered *** ****** ** ********.

*** *******, *****' *****/**** **** ******** and ******** **** ********* *** ******:

IPVM Image

**** **** *** ** *********** ****** when *** ****** ***** ******* *****, shown ***** (**** **** ******* **** not ********* *** ****** **** *****). The ****** ******** *** *** ******** behind *** ******* ** **** ** the *********** ******, ********** ** *****.

IPVM Image

No ** *** **** ***********

******* ******** ***** ** **** *** measurement. *** *** **** *** ****** temperature ** *** ** ****** ********, shown *****:

Single ******* *********

*******'* ********* ***** ******** ******** ** the ***** **************:

IPVM Image

*******, ** *** *******, *********** *********** jumped ******* ******** ********, ******** ********.

Little ****** **** ********

** *** *******, ******** *** ****** effect ** *********** ************. ******* ********** 1-3m ******** (~*.*-**') *** *********, **** subjects ******** ** *** **** ******** used *** ***********.

** *** *****, *********** ********* **** ~0.2°F **** ** ** **:

IPVM Image

Grekkom *********** *** *****

*** ******* *** *** **** * single ******* *** ***********, ***** **** name "***********." *********** ** ******** ***** a ****** **** ****** **** "***" to "****", **** ** ***** ********** of *** ***** ******** ****** *********** measurement.

IPVM Image

*******'* ************* ******** **** **** ******* should ** *** ** ********' **** is ******* ** *** ****** (***** below), ******* ********* ********** ** ******** or ***** *******.

IPVM Image

** ****** *** *********** ******* ** this *****:

******* ********** **** *********** ** ********* at *** ******** ***** **** ** screened *** **** ** ******** *** expected ** **** ***** **** ********, masks ** **** ****** ***********.

Axis ******* ** ******* ****

**** ******** **** **** ** *** market *** ******** *** **** ******* and ** *** ******* ** ********* any ********** ******* ***********.

***** ******** **********, **** **** ******* cameras *** ******* ** ******* *********** measurements, *** **** *** *** ******** by **** *** *** ******** ********* of ***** ***** ********* *** *** diagnosis, ********** ** ********** ** ******* or ****** **********. ****, **** ******* cameras ***, *** *******, *** ******** by *** ** *** ** *** this *** *** **** ** *** specifically ****** *** ** *** ******* products *** *** ******* ** ***** Body *********** ***********.

**** *** ** ********* ****** ******* of *********** ******** ******* ************* ** numerous ************ ***** **** ******** *** we *** ***** **** **** ** these ******** **** ********* ********* **** together **** **** ******** **** ******** for ***** **** *********** ***********. **** of ***** ********* **** ******* **** validated ** **** **** ** ** not ******* ** ********* *** *** particular ******* ** ********.

Other **** ******* *********** *********** ****

** ******** ** *******, ***** *** other ********** ***** **** *******, *********:

** ******* *** ** *** ** these ********** ** ******* ** *** testing, *** ******* ******** *** ******* informed ** **** ***** *********** *********** solution ** *** ****** ** **** app *** ******** ******** ********, ** well. ******* *** *** ********* ** our *********.

*******

*******'* *** *** *** ** **** of $*,*** *** *** *******. *** Axis *****-* ***** ****** *** ~$*,***, depending ** ****. **********, *** ****** and ******* ******* (~$*,*** *****) *** priced ******** ***** **** ***** ******* with *********, **** ** ***** ** Hikvision (**** ~$*,***).

Comments (8)

Another bad joke in the halls of mis-applied and mis-marketed thermal comedians.

It's not only not funny, it's sickening.

Agree: 8
Disagree: 2
Informative
Unhelpful
Funny: 1

**UPDATE 6/11/20** Grekkom has contacted us asking us to re-test their HTC application as they say the version is out of date and the results are invalid. Per their response below:

Hi John,

First of all I want to thank you for taking the time to do an in-depth review of our analytic. We are sorry to see that some of your testing results were not as satisfactory as expected, however I believe that this is due to a combination of unfortunate reasons.

As you know due to the present situation there is a huge demand of these types of solutions, which means that our personnel is a little bit saturated and the review process you were performing slipped through our radar. Because of this, not only we couldn't take the time to guide you through the correct use of our analytic, but also we provided you with an outdated version of it.

In our latest stable version we have some of the issues mentioned in the review fixed, such as the lack of prompt for recalibration when the readings go off. Also there is a number of fixes that contribute to a better stability and detection precision. We also included a number of display options to customize the way the user receives the results.

Another aspect that contributed to some of the not so positive aspects of the review is the fact of that there seems to be some confusion about the analytic features we published such as multiple simultaneous detections and outdoor detections, these features are only available for our server based solution, the edge based one only admits one person at a time and is always recommend for indoor scenarios.

The result that worries us the most is the comment about the precision, we performed extensive testing with thousands of people and in several countries with different conditions, ethnics etc... Before launching the solution with AXIS Q2901-e, we delivered a demo license to almost every AXIS office worldwide and we never got that poor level of accuracy when correctly calibrated. At this point our analytic is being used by a good number of clients who are also quite happy with the results once they learn how to work with it. I'm sure that you observed those anomalous values but I wonder if it was due to a misunderstanding with the calibration procedure.

For these reasons we are wondering if you would be opened to extend your testing and add new conclusions with our most recent version and with the help of one of our technicians to guide you through he calibration process and address your questions.

Kind regards

However, during testing, we contacted Grekkom with questions and examples based off our findings, so they were fully aware and informed of progress throughout. Additionally, a Grekkom employee remotely installed and demonstrated the application for us, so any version or update discrepancies were no fault of our own.

We will update the report accordingly once re-testing is complete.

Agree
Disagree
Informative: 1
Unhelpful
Funny

Dahua 4000 with BB tripods, NVR 16CH and supports.

Regards

Agree
Disagree
Informative
Unhelpful: 2
Funny: 1

Did they send you the improved software yet?

Agree
Disagree
Informative
Unhelpful
Funny

Hello UI3,

We have received the updated application firmware, and will provide an update either later this week or next week.

Thanks!

Agree
Disagree
Informative
Unhelpful
Funny

I'm using this set up in 2 receptions and having really mixed results as well.

I had support on a remote session with an issue of non detection and I was told they were going to upgrade the plugin. So far no update. I would be very interested in your update, and I'll let you know mine :)

Agree
Disagree
Informative
Unhelpful
Funny

Any time frame on testing the updated firmware?

Agree
Disagree
Informative
Unhelpful
Funny

Hello UI #5, we have tested V1.5 of the Grekkom HTC app, and here is what we found during out testing:

UPDATE: Grekkom HTC Version 1.5 Tested

IPVM has tested version 1.5 of the Grekkom HTC app. The core temperature measurement process is the same, with Grekkom using a weighted average and calibration needing to be done when users notice a significant difference in measured temperature. Misses and other issues we've already noted are still present, such as measurement on hands/arms, influence of external temperatures if aimed outside, and hats/masks/hair blocking elevated skin temperatures.

We have noticed that on a subject with a 99.0°F temperature, the HTC application stayed within ±0.4°F of that subject temp, which is more stable than what we had seen previously.

However, the other issues still remain, and the only other major changes have been to the HTC overview controls, allowing for things such as an admin lockout, audible tone mute/unmute, and a toggle for showing/hiding the temperature measurement and area.

Agree
Disagree
Informative
Unhelpful
Funny
Subscribe to IPVM Research to read the full report.
Why do I need to subscribe?
The IPVM Research Service includes products tests and shootouts plus competitive and financial analysis, helping decision-makers better evaluate purchasing, partnering, developing, and/or competing against companies in physical security.
Already have an account?
Loading Related Reports