Gun Detection Analytics Are Here, Says Startup

Author: Brian Rhodes, Published on Jul 03, 2014

A timely innovation, or cashing in on tragedies?

A new video analytics company is touting its 'gun detection' analytics that can trigger as soon as a weapon is drawn, potentially alerting authorities in seconds and saving lives.

Are they crazy, or will this be a powerful defense against active shooters and lower the risk of gun violence? We spoke with the company and take a look at its features, pricing and positioning inside.

* ****** **********, ** ******* ** ** *********?

* *** ***** ********* ******* ** ******* *** '*** *********' analytics **** *** ******* ** **** ** * ****** ** drawn, *********** ******** *********** ** ******* *** ****** *****.

*** **** *****, ** **** **** ** * ******** ******* against ****** ******** *** ***** *** **** ** *** ********? We ***** **** *** ******* *** **** * **** ** its ********, ******* *** *********** ******.

[***************]

Overview ******

*** ***** **** ***** ************ ********'* ******* ********* ********. *** red ***** ***** ********* *** ******** ** *** *** **** the ******, *** ******** ***** ** *** *** *******, *** the ********** ******** **** * ******* ** **********:

********'* ************,*********, ****** *** ******** ****** ********** ** *** ****; ********, rifles, ** ******** *****. **** ***** ***** ** *** **** demo ** '******* ******' ******** ** *** ******** ** *** previous ****:

Key **************

*** ******* *** ******** ************ ** *** ******* ******** **** SafeZone. ** * *******, ******* **** ****:

  • **** ** ****** ********** (** *** ******)
  • ***** ** *******
  • ******* ********* ******
  • ***** *** ********: *** ******* ********* *** *** ** **** as "*********, *** **** ** ******** ** ********** ****** *****"
  • *** ***** *********** ******* (*** ** **** ************* *****)
  • ******* *** ********* ******* ** * ****** ** ** **** above ****** (******** '*** ********* *****, *** ******* *****'.)

******** ******* ** * ********** ****** (****** ********) ********** '************ GPUs' **** *** ****** ** ** *** *** ***** **********. At ******* ********************, *** ******** *** ******* ** ******* ** **** **********.

*******

******** ** ********* ******** ** ******** ************. *** ******* ********** **** ***** ******* ******* ** ******** needs *** **** ***********, *** ****** **** ** ~$**** *** camera *** **** ***, *** ************ $*** *** **** *** software ***********.

******* ******** ******** *** ******** ******** ********, ********** ********, ************* and *********** **********. 

Secret *****?

***** *** **** ******, ** ******* *** ****** ** **** how ******** ***** ******** *** ** * *****, *** ***** alarming ** ****** ***** '**** * **********' *** *******. ******** responded ****:

"[** ***] * ****** ********* **************-***** ****** **** ********** ***** primarily, *** ***** ** **** ******** **. *** ****** ******* classifier *** ********* "***" *** ********** ******* * ********** *** an **."

*** ******* **** ******* *** ******** **** "******** *********** *** **** **** *** ********" *** ******** "*** **** *********** ** ****** ******* ******** **** *****".

Buying ********

********* ** *** ********* ****** ** ********. *** *****, *******, configuration, *** ******* **** **** **** *** *******, **** ** plans ** ****** ** ******* *******. *** ******* ****** **** is ** **** ******* ** *** ************* *** ************* ******** for **** *****, **********:

"******** *********** *** ************ ******* **** ******** ********* ********** *** each ******. ******* **** ** ******** ***** ********, ********, *** optics *** ******** ** *****-******** *************** **** ******** ** ******** to ****** **."

**** **** **** ****** ******** ******* ** ************* ***** ** needed, *** **** **** **** * ********'* ******** ************ ****** if ******. ******** **** **** ********* ******* "***** **** ********** **** **** ******* **** ********, ****** *** the *********** ** **** ******, *** * **** ** *-**** addresses *** **** ******* ******* ** ** ******** **** ******** detects * ******."

How **** ******?

********* ** ***** ** **********, **** **** ** **** ***? While *** ****** ** ************ **** ***** ** ******** ** *** system ********* ********** *** ******* ** ** ****** *******, ** does *** ********* ******** **** ** *** ****** ** **** effective **** ******* ********.

******, ** *** ******* ******* *** * ******* *****, ****** on ******* **** **** ****** ***** ****** ******* ** ****** ************. ** **** point, *** ***** *********** *** *** ****** ** *** ***** **** ****** ** a ******** ********* ** ******* **** ***** *********.

***** **********

** **** **********, **** ******* ***** *** **** ** ********* is ****, **** *** ****** ******. *** ********* **** * **** gun ** ****** ** ***** ** * ******** ******* ****** is ****** **** *** - **** * ****, ******* ******.

** ********, **** ** *** ****** ** ********* ******** *** only ***** ****** **** * ****, **** ***** ** ** ***** alarms *** ***** ***** *** - * ***** **** ***** cause **** ********* ** **** ***** ** *** *********.

**** ** * *********** ***** *** *** ******** **** ******** to ******** ********* **** ******.

Missed ****

***** **** **** ** ********, ****** *** ****** *** ****** *********** *** **** **** objects **** **** (*******, *****, *** ******, ***.), ** ** also * ******* ** *** ********* **** ** ******** * real ***** ***.

Going ** **** ********

** *****, ******, *** ********* ** *** ** *** ***** hopes *** ******** ********, ** ** ****** **** *** ******* will **** ******** *** ******* *********.

 

Comments (11)

What about pop-tarts chewed into the shape of a gun?

Or, just pointing out a pretty bird to a class-mate? Finger-pointing constitutes an assault weapon these days.

I can only imagine the number of false alarms this will produce.

Given that trained police officers often have a hard time distinguishing a firearm, I am suspect about this system.

In the given demos, there are no other people around, giving a clear view.

I'll put on my 'companies are buying the government' tin foil hat and say that they'll advertise this, get it on the news, lobby congress to require it, and make a @#$% ton of money.

Next time anyone draws a gun and shoots up a place, we will see the media and the family's declaring that this would have saved them.

At $4,000 per camera, a college campus with 1000 cameras would need $4 million to start with. I see people taking cameras down rather than being forced to have that expense.

Even if this product works 100% as advertised, it is difficult seeing the value for the expense.

If a guy is brandishing a gun out in the open, he is most likely seconds away from using it. At best this system warns me seconds ahead of time? ...that's not much early response to be gained by this analytic.

With the substantial analysis of recent mass-shootings, I'm not sure I have heard anyone claim the problem is instant notification and awareness of the shooter.

Even if this product works 100% as advertised, it is difficult seeing the value for the expense.

If it works 100% as advertised, it is actually easy to see the value for the expense. NONE!

Can anyone explain what they are even claiming on their critical, make-or-break page, titled "How Does SafeZone Save Lives?".

This is a recreation of the page, my comments in <italics>

How Does SafeZone Save Lives?

What would happen today if an armed gunman or gunmen attacked a building where you are, or where your children are, or where your employees are? Would an effective response be launched and prevent the loss of lives?

<The implication is no. Note: the promise of an Automatic Threat Response by the title of the picture>

Here is a video of a simulated gunman attack on a public meeting space

In this video clip, the gunman is present for approximately 60 seconds. If this were a real attack with a full room of people, dozens of people would have been hurt.

<Agreed. But what can be done to save them?>

So how does this scenario change with SafeZone?

Here are the results possible with SafeZone detection and response support

The threat is detected and a response is launched in 10 seconds, before an attack can be mounted. <I see no response launched! Or if one is it's certainly not stopping the mounting of the attack, which continues in all windows unabated!> Responders would have a chance to engage the gunman, potential victims would be alerted of the gunman's presence, and automatic defensive systems (such as door locks) could be engaged.

<Were some doors locked, somewhere? What am I missing?>

Well, that's just in that there scenario! They have more analytics that regular folk might find handy:

I'm sure y'all had this one happen to you: You just finish installing a ski slalom pole in the middle of your front, corner view yard in July, and then your onry neighbor has the nerve to walk right up to it in broad daylight and graffiti all over it. But that never has to happen again...

It looks easy nuff to detect graffiti in broad day light 10 feet away from you with a straight on FOV, but have another gander, the scene is a might bit tricky. Watch how the analytical thing will not be distracted by the possible garbage digger (who doesn't seem to mind the graffitist) nor by the seasick tilt of the whole apparently "crooked" street.

Disclaimer: I am being facetious here, and am really not impressed with this at all. Are y'all?

Did the company itself give an estimated false/missed alarm percentage. If they refuse to give one, or say it's 100% accurate, you know that it's probably bunk.

Watch the rifle detection video. It appears as if the analytic will only detect an object as a rifle if it is held in a firing position (parallel to the ground). And even then, it doesn't maintain a lock on the rifle once it has been detected as such, reacquiring, then losing the target box continuously as the shooter moves down range.

Below is another video from these guys showing how their analytic can tell that a broom isn't a rifle. But it has problems telling that a rifle is a rifle too... :)

Notice the difficulty the analytic has in detecting (and reacquiring) the actual rifle unless the rifle is held (in a firing position) perpendicular to the camera... i.e. if the rifle is pointed at the camera (or in any direction except sideways to the camera) the analytic won't recognize it as a rifle.

So, some dudes got together and catalogued pictures of known firearms (from a side view perspective only apparently), built an analytic to look for these images in camera FOVs, and claim they have a firearm detector that is worth $4K/camera.

As Brian already mentioned, even if it worked flawlessly and immediately identified every firearm in a FOV, what is the 'value' of such a thing?

Keep in mind that any funds spent on these types of gimmick (imo) solutions are not being spent on other security 'things' that actually possess value. So, a case can be made that 'solutions' such as these actually come with a negative value.

...so we expect that the product will draw interest and perhaps heartache.

Why heartache, because they will deploy it and a situation will ensue where it was needed but failed to 'perform'? More likely would be deployment followed by false broom alarms leading to heartburn.

Liked the "Going to Draw Interest" bit.

Pun Score: B+

This is the Redeemed Rambo deployed by helicopter in Southeast Asia, fighting the good fight by Uzing any and every means to bring the boys back home:

And Ari and I* would appreciate his good image not be sullied by yet another association with an agressively stupid active shooter product.

*Possibly Marty also, though he might have just been messin' with us.

Now hold on just a sec, before you BRate BRhodes, you might take a lil look see at yonder rectangle. On account of it not being all red boxed, like all the other clearly contrasted guns are. That means it identifies the weapon, but has also id's the possessor!

And that new-fangled Rambolytic detector is thrilled to see Mr. Balboa give somebody a buzz cut or even help move a bullet-proofed bookshelf in front of a grammar school door.

You see the Rock ain't the disease, he's the cure...

Login to read this IPVM report.
Why do I need to log in?
IPVM conducts unique testing and research funded by member's payments enabling us to offer the most independent, accurate and in-depth information.

Related Reports

Genetec Privacy Protector Tested on Nov 12, 2018
Genetec has built in Kiwi Security's Privacy Protector directly into Security Center, an analytic which anonymizes individuals in cameras' fields...
Kogniz Silicon Valley AI Startup Profile on Nov 07, 2018
Kogniz is a Silicon Valley company that aims to bring AI analytics to security and surveillance, centering on their own smart cameras: We spoke...
Avigilon Opens Up Analytics And Cameras on Nov 06, 2018
Avigilon is opening up. The company historically famous for advocating its own end-to-end solutions and making it harder for 3rd parties to...
Directory of Access Control and Intrusion Startups on Oct 30, 2018
This directory catalogs access control and intrusion detection startups IPVM has found noteworthy. The directory is provided for your review to...
Exacq Co-Founders Return, Start Qumulex on Oct 24, 2018
Exacq co-founders Dan Rittman, Tom Buckley, and David Underwood are back, starting up Qumulex. In 2000, they sold Integral to Andover...
Startup SafePass Profile on Oct 19, 2018
A major problem with visitor management is that the systems mostly require adhesive printed paper labels and paper logs, creating waste and an...
Huawei Admits AI "Bubble" on Oct 16, 2018
A fascinating article from the Chinese government's Global Times: Huawei’s AI ambition to reshape industries. While the Global Times talks about...
Dahua Face Recognition Camera Tested on Oct 15, 2018
Dahua has been one of the industry's most vocal proponents of the value that AI creates: As part of this, Dahua has released a facial...
IACP 2018 Police Show Final Report on Oct 08, 2018
IPVM went to Orlando to cover the 2018 IACP conference, the country's largest police show (about as big as ASIS), examining the 700+...
Anti-Tailgating Startup: Spyfloor on Oct 03, 2018
A Canadian startup, Spyfloor, is using a different approach to warn against tailgating, a common access control problem. By counting feet,...

Most Recent Industry Reports

Magos Radar Company Profile on Nov 12, 2018
Magos America General Manager Yaron Zussman admits when he first came across Magos, he asked himself: "What's innovative about radar?" Be that as...
Genetec Privacy Protector Tested on Nov 12, 2018
Genetec has built Kiwi Security's Privacy Protector into Security Center, an analytic which anonymizes individuals in cameras' fields of view...
Chinese Government Increases Hikvision Ownership on Nov 12, 2018
The Chinese government - Hikvision's controlling shareholder - is increasing its ownership of the video surveillance giant amid sharp stock price...
Axis: "No One Wants To Buy A Camera" on Nov 09, 2018
Axis has, in its own description, made a bold declaration: The industry is changing so rapidly that the following statement might seem bold but...
Video Surveillance Hard Drive Size Statistics 2018 on Nov 08, 2018
What is the most common hard drive size for video surveillance? 150+ integrators answered: What size hard drive do you most commonly use? What...
Axis 2N Intercom Tested on Nov 08, 2018
Axis expanded its video intercom business buying Czech-based 2N in 2016. Despite competing against owner Axis' intercoms, 2N recently registered as...
Haven Targets School Security with Lockdown Lineup on Nov 08, 2018
Haven, a US startup founded in 2014 as a residential-focused company, has now raised funding and is offering a lineup of commercial grade locks for...
Ubiquiti Protect Video Surveillance Profile on Nov 07, 2018
Ubiquiti has now been in the video surveillance market for 7 years (see our first coverage back in 2011). In that time, the company's revenue has...
Kogniz Silicon Valley AI Startup Profile on Nov 07, 2018
Kogniz is a Silicon Valley company that aims to bring AI analytics to security and surveillance, centering on their own smart cameras: We spoke...
Dahua Dual Imager Dome Camera Tested (HDBW4231FN-E2-M) on Nov 07, 2018
Dahua has introduced a dual-imager dome model, the HDBW4231FN-E2-M, with two independently positionable sensors including integrated IR, not found...

The world's leading video surveillance information source, IPVM provides the best reporting, testing and training for 10,000+ members globally. Dedicated to independent and objective information, we uniquely refuse any and all advertisements, sponsorship and consulting from manufacturers.

About | FAQ | Contact