Testing Color Analytics Performance

Author: Ethan Ace, Published on Dec 16, 2013

In an industry known for overhyping its own value, color analytics have been considered pure science fiction, with few, if any, stories from the field to back up manufacturer claims.

In this report, we test Bosch's IVA color analytics to see which practical applications they performed in, and where they failed.

First, in an interior conference room with a human subject:

And an outdoor parking lot/driveway with different colored vehicles:

We varied the precision of the analytics to understand how performance varied, from least to most sensitive using their built-in scale:

We then tested a variety of colors from red, green, blue, orange, neons, etc.

**** ************* *** ********** *** *** *****, ***** ********* **** **** considered **** ******* *******, **** ***, ** ***, ******* **** the ***** ** **** ** ************ ******.

** **** ******, ** *********'* *** ***** *********** *** ***** ********* ************ **** ********* **, *** ***** they ******.

*****, ** ** ******** ********** **** **** * ***** *******:

*** ** ******* ******* ***/******** **** ********* ******* ********:

** ****** *** ********* ** *** ********* ** ********** *** performance ******, **** ***** ** **** ********* ***** ***** *****-** scale:

** **** ****** * ******* ** ****** **** ***, *****, blue, ******, *****, ***.

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

Key ********

  • ** ****** *** *****, ******* *** ****** ** ******* **** the ***** ***, ****** ******** ********* ** ****** *** **** difficult. **** *******, ******* *****, **** **** ** ****** ****.
  • ******* ********* *** *** ******* ** ****** ******** ** *** desired *** ** *** ***** ***** ********** ****** (*.*., ****** and **** ********** **** **** ** *******).
  • ******************* ** * (********), **** *** **** ***** *******, ******* the ******** ** ***** ******* **** ***** ******, *** **** increases ****** ** ****, ** ****** **** ** **** ******** in ***** ** *******.
  • ******************* ** * ** * (**** ** **** ****) ******** in ******* ******* ** ****** ******* (**% *** **) ** alerts *** ********* **** ** * ***** ***** ** ****.
  • ** *** *****, ********* **** ****** ******** ******* ** * nearly ***% ******* ** ******, ** *** ****** ****** ******* colors ********, ****** *** ***** *** ***.
  • ** * ****** ******* ******* *** *****, ******** **** ************ detected **** ***** ********* *. ** ********* * *** *****, rules ****** ** ******** ** ***. 

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

***** *** ********* ******* ******* ******** *** ***** ******, ** do *** ********* ***** ********* ** **** ** ******** ************. Alerting ** **** ******** ****** **** ****** ** **** ****** matches *** ** ****** *********** ** *****, ***, ** ******** lighting *** *******. *******, ******** ********* ******* ** ******** ****** for ****** ***** **** **** *** ** ********. ******* ** this, ***** ********* ****** **** ** **** **** ******* ** reduce *** ****** ** ****** **** ** **** ******** ************.

It ***** ** *****

*** ***** ********* **** **** ***** ** ** * **** end ****** (*** ***** ***-****) **** *** ***** **** ** *** ***** ******** **** *** *** ****** *** ***** ***********. *********** ***** ****** ** **** ***** ** ***** ******* cameras *** **** **** ****** ***** ***** ********* **** **** to **** **** ******* ******* *** ****** ****** ***********.

Configuring ******

**** ********** ******* ************* ** ***** ********* ** **** ** some ****** ****** ** *** **** ***** ********:

Selecting ******

*** ******* ******** ** *********** *****'* ***** ********* ******** ** compensating *** *** ********** ******* **** *** ****** "****" *** the ***** ******** *** *** ***** *****. *** *******, ** this *****, *** ****** ******* *** ** *** **** *******, as **** ** *** ****** ********** ** *** ***** *****, and ******** *** ********* ****:

*******, ** ***** *****, *** ****** *** ** ***, ****** mostly **** *******, ******* ** ******:

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

  • *** ****** ** ****** *** *******'* ***** ** *** ******** color, ***** ** ****** ******.
  • ** ***** *****, ********** *** ***** ***** *** ***** **** the ****** ********** ** ** ****** ****** *** **** ******.

*******, **** ******* *** ******, ******* ********** ** *** ***** of *** ********** ****, *** ****** ***** ******** **** ****** as ****** ****, ********** ** ****** ***. **** *** **** of *** **** ***, *** ****, *****, ******, ******, ***.

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

** *** *****, ********* *** *** **** ********* **** **** colors **** ** *****, ****, *** *****. ****** **** *** simply ******* *** ****** ***** *** *** ***** ** *** same ******** **** ** *** ***** *****, ****** *** ** variances ** ********** **** ******, ***.

*** *** **** ****** *** ******** ** *** ***** ***** also *** **** ******** ** ******, ** **** *****. ****** the ******** ****** ***** ***** ********* ***** ** *** ****** of *** ***** ******* ** **** ** *****. *******, ** also *********** ******* *** ****** ******* ** **** ***** **** the ****** ** *** *****, ****** ********** ********** **** **** difficult.

Color ********* **********

*** ***** ********* ******* ****** ******** **** ** ** ******** down ** **** * *** ******, ******** ** ******** ********. This ******* ****** **** * (*** ***** *******) ** * (only *** ******** *** *******), ** **** *****:

** *** *****, ** ***** **** ********* ***** * ******* reduced ******** *****. *******, ** **** *****, ****** ***** **** the ******** *** *** ******* ******, ** ****** *** ****** may ******* **** *** ** ********, *** *******. ** ********* 2 *** *****, *******, ******** *** ************, **** ***** *** rates **** **** ********* *, ** *** **** ********* ** all, ** **** ** ******** *****.

Color ********

** ***** ** * **** *** ****, ***** *** ***. Our ******* ****** ****** *** **** *** *****, ********* ** the ********* *******. ***** ***** **** ************ ********** ********** ** color, ********** ** ********* *, ***** ********* ** ***% ** attempts.

*****, ***, *** **** *** *****************, **** ***% ******** ** ********* *, ******** ** **% at ********* *. ****** ****, ** ********** * *** *, exact ****** ****** *** ******** ** ******* *** ****:

****** **** ****** **** ** *** ****** *** ***** ******** below **** ********* ** ***% **** ********* *, *** *** not ******* ** *** ** ********* * ** *****. **** was ******* *** ****** ***** *** ***** ****** ** ****** grey ** *****, ***** *** ******** ** ****** ********* ******.

Low *****

******** ***** ** ***** * ***, ******** ** **** ********* 1 ***** ************* ** ****** *** *****. ** **** ***** level, ** **** ***** ********* *** ********** ****** ** ********** for ****** ********* **** ** *** *****, *** **** *** no ****** ** *********** **********.

*****, ***** ***** ** ***** **% ******** ** ********* *, and ** ********** ** * *** *****:

 

********* ** *** ******* ****, **** ***** ***** ** ********* 1 ******** ** **** **%, **** ** ******** ** * or *****:

**** ********* ****** **** ****** *** ** *****, *********** ***** 70% ********, ****** ***** ******* ** ***** ****** ********* *:

**** ***** ********* **** *** ** *** ****** ** *** light, *** ** *** ******* **********. ** ********* *, ** remained ***% ********. ********** ** ********* *, ** ********** * 30% ***** ****, ****** **** *** ***** ***** ******:

*******, ****** *** ** ******** **** ****** **% ** ********* 1 ** *** *****, *** *** *** ******** ** * or *****: 

Vehicle ********* *********** 

** ********* *, ******** **** ************ ******** *** ** ***** larger ****. *******, *** ** *** *******'* ********, *** ***** changes ** ********* ******* *** ********* *** ******* ****, ****** the **** ****** ***** ****** ** ********* * *** ***** fail ** *******. **** *** **** ***** ******** ****** ** vehicle, ********* ****:

*** ***:

Comments (19)

*** **** **** ** ********* ** **** ******* **** ******* units ********** * ********* ******?

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

**, ****** *** *** *********** ******* ******* ****** *** ******** temperature. *** "********" ******* ******* *** ***** ************ *** ************, not ***********: **** ***** **** **** ******* *** **** ****** with ********* ************, *** *** ***'* **** ** **** *********** is **° ** **°.. *** **** **** **** ** ** different, *** *** ***** ****** (***** *** **** **** *********; always *****/***** ****) *** ********** ******** *** ***** ***** ** frame ** *** ****** *********** *** ***** ***** ********* ** the ********* ************ ** *** ********* ****** ******* ** *** scene ** **** ******.

**** ***** **** ***** ****** *** **** **** **** ******** to ****** *** ***** **** **** *** ****** * ***** in ** ******* ******.

******,

******

(***********)

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

/******

*******!

***** *** ** **** **** *** **** ***** ******!

* **** ******* ** ** *** ****** ******* **** * "fight" ***** ****** ** ******** **** *** **** ** ** difficult ** ******** "*****" ** * ******** ******* *** *********! And ** *** *** ******* ** ******* *** ******** ****** to ******** ********** ********* (*** **** **** **** ** ******!..) without ****** ********* ** *** ********* ******** ******..

**** * ***** ****,

******

(***********)

***** *** ***** & ****

***** ****** , **** ****** *** *********** ******* ** *** color ******** ** **** ******* .

***** ***

******* *** ** ***** **** ** * "*******" ********?

**: ******** ****, ******* *** ********/****, *******, ***. ********* ** you *** * ******* **** ******* ******** *** ***** ***.

*****, ******* *** ********/**** *** ******** *** ***** ** ** especially *********** ** ******* ***** ****** *** ****** ************ *** going ** ****** *** ***** ********** ** ******* **** ****.

* *** ******** ** ******** ************ **** **** **** "*******" vs "**** *****" ************.

***** *** ***** ******* ******** *** *** "*****" ** *** times, ******** **** *** ****** (** **********).

***, ** ** *** *** ******* ****** '*********' ** ****** work ****. *** ******* (* ** ****** **** ** *** I ******* **** ***** **** ****), *** **** * ******** carrying ***** *** *** **** ** ************* ****** *** **** fruit **** '***' ******. ** **** * ********, *** ***** carefully *** ** ** *** ******** ** ** ********** ******** those *** ***** ******.

***, **** ***** ** * **** *******.

** ** * "*******" ******** ** ****** **** * **** % ** ******** ********* ***'* ******...

* ***** ****** ** ***** **** ** **** ** ***** test *** *********** ****** ******* ** ;-)

*****, ******** *** ***** **'* ******** **** *** *** *****, about ****** ***** ******.. ******** **'* *** **** ******* ** this ********* ******.. *** ******* ** **** *** "*****" ********** always ***** ***** ** ***** ***** ** ******* **********, **** perfect ****** **********, *** *** *******.... *** ************* ***** **** is *** *** *******.. ***** ****'* **** * *****, ******* unfortunately ** *** **** ***** **** *** ********** *** ******** never **** ****** *** **** ****..

*** ***** ******, *** ********* ****** ** **** **** **** it's *** ****, **** ******* ******* *****, *** **** **** wants ** *** ********* *******....... ** ****** ********* ***** *** in *** ******* *** **** **** ********; ** "*****" *********** of ****** **** ** **** ** **** ********.. *** ***** it's *** **** "*****" *** ******* ** ** ** **** way..

***** *** *** ******* ***** **** **** ******** ******* ** fixed ********** *********.. *** ******* **, ***** ** ** ****** method ** *** ** *** **** * ******* ** ****** perfectly ****** *** ********** ***********, ** *******, ** ***********, ******* stable *****.. **, *** *** ***** *** **** ** * person ** **** ******!.... *** *** * *******... ...*****! **** trick..

** * ****** *** ** *** ******** *** **** ** me "** *** **** *** ***, * *** ** ** my ****!....": ***, ** *** ********* ******* * *** ** a **** ** * **** ******** *** * *** ** my **** * ***** ******* ** *...... *** **** *****'* mean *** ***** *** ****** ******* ** *, *****?....;)

******,

******

*** *** ** ;-)

***** ****'* *** ** *** *** **** (******* ****), ** appreciate *** **** **** **** ***** ** ******** *** ***** screens *** ********* *******...

** *** ***** *** ***** ****** ****** ************ ** *** ******** *****. ** *** **** * *** ** ******** * ******* person.

***** *** ***** ******** *** **** ***** *********** ***********. **** to ********* ** **** ******, **********, ********, *** **** ****** you *** ***** **** **?

****, ******* *** ****** *** *** ********** ****** ** '******** will ** ***** ** **** ******' ***** ***** *** "**** Without ********** **** ****" ********.

**** ****** **** ***** **** *** ***** *********, *** **** can *** ******* ***** ********?

***'* ****? ***** ** **** ** *** ***** ** ***** then *** ****** ** *** ****** **** *** *** ********* after *** ****?

*** **** **** **** **** ******** ** ** ***** ***** it's ********* ** ***/****/*****/*** ***** **** **'* ********, ** *** video ** ***** ** *** ****?

* ***** **** ** ***** **** ********* ***** ** **** powerful ***** *** ***-******** ** **-********, **?

**** **** ******* ***** *** *** *** ***** ***** ** in, ***** *** ******* *** **** ** *** *** ********* against *** ******** *****...

***, *** "******** ********"...... ** ******, ******* *** ****** ** color **** **** **** **** ** ***-**** ***** ******* ** a **** *** ********* *** ******, *****?.. ***, ***** **** disk!....

** ***** ** *** ******* **** **** **** *** ***-**** video ** * ***** ******, **** ** ** ***** **** bright *** ***** *** ******* ** *** ****** **** **** to ******, **** **** *** *** ********* ********* *** *** color *** *** "**! ****'* **!!..."........;)

*** ******* ********* *** ****** ** **** ** **** *** need ** ** ** ******** ** ******. *** *******, ** I ****** *** * *** *** *** *** ** ******* and **** *** ** ****** ***, * *** ******* **** through *** ** ********** ****** **** **** *** **** ***** over *****. **** ** ****** ** ******** ** *** **** that *** **** *** *** ** *** ****** ********.

*** ****'* ******** *** ***** ****. ** *** ********* ******, the **** ******* ** ** *** *** ***** *** **** it.. **** ** ****** ** *** ******, *** ** *** then ******* **** ** *******: **** ***** *** *******, ****** over/under-exposures, **** ****** ***** ************* *** ********* *****, *********** ** sun ** *** *****, **** ********** ******, ** *****, ...

* ********** ***'* *** **'* *******, *** ** ***. ***** are **** ******** ** ******** ******** ************, ***** * ****** timely *********** *********** ** *** *********** ** ** *** *********** itself.. ****** **** ** ** **** **** ***** ********* ********. Only, ** ***** ** *** ****** *****'* * *** ** science ******* *** **** **** ** ******* **** ** **** this ****** **** ********* *** ***** ************..

******, ******

******

(***********)

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

Hanwha Wisenet X Tested on Mar 27, 2017
Hanwha has released their latest generation, the Wisenet X series, powered by their new Wisenet 5 processor. This new series claims improvements...
ObjectVideo Acquired by Alarm.com on Mar 14, 2017
Once potentially the most recognized name in video analytics, ObjectVideo has effectively been sold for pieces. Alarm.com has now acquired what...
Hikvision People Counting Tested on Mar 09, 2017
People counting has historically been a difficult task, requiring specialized cameras, expensive software, and careful camera placement. But in the...
SimpliSafe Camera Tested on Mar 07, 2017
SimpliSafe ia one of the most controversial companies in the industry, as they have become the symbol of the DIY threat to traditional alarm...
Positive Video Surveillance Outlook - 2017 Statistics on Mar 07, 2017
376 manufacturers and integrators revealed their outlook on the future of the video surveillance industry in a recent IPVM survey. Both groups are...
Vivotek 3MP 180 Wall Mount Camera Tested on Feb 28, 2017
Purpose-built wall mount panoramics have become an attractive niche to cover entranceways. In 2013, we bought and tested Vivotek's 1MP...
Uniview (UNV) IP Cameras Tested on Feb 22, 2017
"We're #3," in China says Uniview (UNV). While the company significantly trails Hikvision and Dahua in total sales, one notable difference is that...
Arecont Vision MicroDome Duo Tested on Feb 14, 2017
Arecont Vision is back with another multi-head camera, this time thinking smaller with the MicroDome Duo, a two-imager model, with the tagline...
Hikvision PanoVu Multi Imager Tested on Feb 08, 2017
Hikvision has entered the multi-imager market with their 180° PanoVu DS-2CD6986F-H, an 8MP, 4-imager model equipped with ~1/1.8" imagers and...
Axis 15MP Multi-Imager Tested (Q3708-PVE) on Feb 07, 2017
Axis continues to expand in the multi imager market, adding to a segment started with the 33MP Q3709. We bought the Q3708-PVE and tested it...

Most Recent Industry Reports

Chinese Government Spies on Churches With Video Surveillance on Mar 27, 2017
The Chinese government is using video surveillance to spy on churches in China, reports UCANews, explaining: The [Chinese government]...
Burglar Alarm Sirens Guide on Mar 27, 2017
Sirens are used to alert users to an alarm condition. In this note, we examine how to choose, locate, and install alarm sirens, including Siren...
Dahua Distributor Bad Breakup on Mar 27, 2017
A Dahua distributor in Europe claims that Dahua took over his top customers, revoked his distributorship status, and left him with a large amount...
Everbridge Mass Notification Service Examined on Mar 24, 2017
Everbridge is expanding in the security space. In January 2017 Everbridge acquired PSIM platform IDV, and have also begun integrating with other...
Hikvision Removing Auto 'Phone Home' on Mar 24, 2017
Facing pressure over their cameras auto phoning home and their Chinese government ownership, Hikvision has begun quietly removing automatic...
Axis Camera Vulnerabilities From Google Researcher Analyzed on Mar 23, 2017
A Google security researcher has reported 6 vulnerabilities for Axis cameras, affecting multiple models and firmware versions. In this report, we...
OpenEye Takes Aim At Exacq on Mar 23, 2017
First Milestone targeted Exacq with a takeover offer, and now OpenEye is gunning for them with an offer to swap out Exacq for their cloud-managed...
Lock Keyways For Access Control Guide on Mar 23, 2017
Lock keyways can be the difference between a lock working or not. Understanding keyways is important for access control. Indeed, a member recently...
Broken Browser Support for Video Surveillance on Mar 22, 2017
Modern web browsers have left the security industry behind. Current Chrome, Firefox, and Microsoft Edge browsers do not support NPAPI plugins,...

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