Avigilon Face Mask Detection Tested

By Zach Segal and Derek Ward, Published Jun 24, 2020, 10:12am EDT (Research)

**** **** ********* **, **** ************ not ******* * **** ****, ** an ********* ******** ** *** ***** of ***********. ******** *** *** ******** theirs *** *** **** **** ** work?

IPVM Image

****** **** ****, ** ***** *** findings ******:

Executive *******

********’* **** ********* ********* ***** **** in ********* *** ********* ***** **** would ** **** (******* **** **** lighting ** *** ***** *** ****** WDR ******). *******, **** ***** ** (for **** *****) ** ** ***** WDR, **** ********* ****** ******.

********'* **** ********* ********* *** ********** with * **** ***** ** ******* including ***** ** *** ** *****, as **** ** ***** ********** ******* used **** ********'* ** *********. ***** is ** ********** **** *** ****** for **** ********

********'* **** ********* ****** **** ****** than*****, ********, *** ****** ******'* **** detection; ************, ********'* ********* ************ ******** of **** ***** ******** **** **** rather **** ********** **********.

Hard ** *****

***********, ********, *** ****** ******'* **** detection,********'* **** ********* *** **** ** spoof. ** ********** * ******* ******** their **** **** * ****, ****** bill, *** **** ** ****-****, ********** an *****.

IPVM Image

***** ** ** *** **** *** these ******* ********* *** ******** ***** analytics, *** ******* ****** **** ********'* training ** ************** *** *** **** robust.

Monitoring **** **** *********

**** ********* ** ********* **** ***** Focus ** ********* ********* **** ******** 'person ******* ****' ******, ** ***** below:

IPVM Image

****** *** ***** *** ** *** up ** “** **** *********” ** triggered, ******* ************* ** ***** ********** the ******.

IPVM Image

**** **** **** * ***** ********* (which ** **** *** ****** *** plan ** ******* ** *** ******) and ** ****** (*** *.*) **** is ********* *** **** ******* **** plans ** *** **** ******-********** ********** and ************ *** **** **** *********.

Performance *********

*** ********* ****** **** **** ** evenly-lit ********, **** ** ***** *** low-lighting (** ****** **** ** ~* lux).

*******, ** *** * *** ********. T**** *** * ********** ********** ** performance ** *** **** **** **. In **, *********** ******** ***** *** be ********** ***** *** ****** *** time ** ******. **** ***** **** mask-less ********, *** ****** ** *** camera **** ***** *** *******.

IPVM Image

*******, ******** ****** ******* ** *** dark ** ******** ** ** * common *** **** *** **** *********.

****, ** ***** *** *********, *** camera (******** *.**-***-***-**) ****** ** ********* mask-less ******** ** ****-**** ** **** instances, *** ** ***** *******. * mask-less ******* *** ****** *** ******* door ** * **** **** *** then *******, ******* ** *** ******, was *** *******.

IPVM Image

******** ********* ** **** ************:

***** ******* ** ***** ******* ******* scenes ** ****-******* ** **** ********

******** ****** ** ****** ******* ******* directly ************* ** *** ******:

*******, *** ********* *** ****** ********** when ****** **** ****** **** ** sharp **° ****** ** *** ******.

Will *** ****** **** **** ******

*** ******** ******** **** *** ******* a "**-****" ***** ** *******'* **** is ****** ******* *** ******. *******, it **** ******** * ***** *****, ~30-degrees ** *** *******, ** *** face ** ***** ** *****.

IPVM Image

Improper **** **** *** ********

** ********, *** ********* ****** ** detect ********** **** ***** (**** *** nose *******), ** ***** *****:

IPVM Image

******** **** *** ****** ***** **** to ** **** *** ** *** be ********* ***** ***** **** ** be ***** **.

Varying ******* ** **** ********* ******

********** ****** *** **** **** * few ******* ** * *** **** of *******, **** *** **** ****** factor ***** *** **** ******* ******* a **** ***** ** *** *** of *** ******. **** *** **** response **** ** ** ******** ** looking ** ********* ***********.

******** ********* ** **** ******:

*** **** *** **** **** ***** for ********** (********** ******, **** ****) until ** ****** ***** *** ***** of **** ******* *** *** ***** periodic ****** ******** *** **********.

What ******** ** ***** On

********'* **** **** ********* ***** ** ******* **** ******** ********** ******™ **********”. **** ******** *** “******** *** *****-***** *****-********* ****** ******* with *** ******** ** *********, ****** patented ****-******** ***** ********* *** ********** Search **** ********* ** *** * or *** * ********”.

PPF ******** *** ******** **** *********

*** ******* ***** **** ** *** was ******** *** **** *********. *** the ****** ***** (******** *.**-***-***-**) ** used, ***** ****** **** ** *** as ~**' **** ** *** ****** focal ****** (*.***) ** ~**' **** at *** **** ********* *** *****.

Firmware ****

** **** ******** ******* *.*.*.**.

Upcoming *******

** *** ********* ******* ******* ******** from ****** *** **** ** ********* those ******** *******. ** *** **** planning ** **** **** **** **** detection ******** *********.

Comments (3)

Great testing, May I suggest you test Herta Security software face recognition even if the subject is wearing a mask as advertised?

Agree
Disagree
Informative: 5
Unhelpful
Funny

What is being detected/not detected when a person is standing with his back to the camera?

Agree
Disagree
Informative
Unhelpful
Funny

No alarm is raised when your back is turned to the camera. If the camera cannot see your face it will not trigger an alert.

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