Milestone LPR Tested

By: Ethan Ace, Published on Mar 31, 2015

How well does Milestone's LPR work?

Milestone's LPR supports off the shelf IP cameras, integrating license plate recognition with the Milestone Smart Client for live video, playback, and alarms.

We tested Milestone's LPR with three cameras including Avigilon's H3 integrated IR bullet, Bosch's Super Low Light Starlight 5MP, and Messoa's specialty license plate capture cam, the LPR606.

The 1 minute video excerpt below, from our 15 minutes of video inside, gives an overview of Milestone's LPR:

The report answers the following key questions:

  • How accurate was Milestone's LPR?
  • What accuracy variance existed across the three cameras tested - Avigilon, Bosch and Messoa?
  • What factors most contributed to missed plates and inaccurate recognition?
  • What setup issues / factors do you need to factor in?
  • What processing load / server impact should you expect?
  • What issues caused missed plates and improper recognition?
  • How does it compare to Genetec's AutoVu LPR?
  • *** **** **** *********'* LPR ****?

    *********'* *** ******** *** the ***** ** *******, integrating ******* ***** *********** with *** ********* ***** ****** for **** *****, ********, and ******.

    ** ****** *********'* *** with ***** ******* ********* ********'* ** ********** ** bullet,*****'* ***** *** ***** Starlight ***, *** ******'* ********* license ************ ***, *** ******.

    *** * ****** ***** excerpt *****, **** *** 15 ******* ** ***** inside, ***** ** ******** of *********'* ***:

    *** ****** ******* *** following *** *********:

    • *** ******** *** *********'* ***?
    • **** ******** ******** ******* across *** ***** ******* tested - ********, ***** and ******?
    • **** ******* **** *********** to ****** ****** *** inaccurate ***********?
    • **** ***** ****** / factors ** *** **** to ****** **?
    • **** ********** **** / server ****** ****** *** expect?
    • **** ****** ****** ****** plates *** ******** ***********?
    • *** **** ** ******* to*******'* ****** ***?

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

    Key ********

    **** *** *** *** findings **** **** ****:

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

    • ********** ** ******* ********* similarly, ********** ********* *** *********** ~95% ** ****** ****** the *** *** ~**% of ****** ** ***** with ******** *** ** 1/250 ****** (****** *******).
    • ******** ********** ** ****** was ****** ** ******* plates ***** ******* */*** exposure *** ** ************ and ******* ** ******* plates.
    • ***-** ******* ********** ********** ***** 94% ** ****** ****** the ***. ** *****, non-IR ******* ******** ********** only ~**% ** ******, as ~**% ** ******* plates **** *** ****** up ** ***, **** only **% ** ***** read ********** **********.
    • *********** ******** *** ******* at ******* ******, **** ~15 ** ~** ***.
    • **** ***** *** **** using ********* *********** ********, <1 ****** ******* ******* passing *** *** ***** generation.

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

    • *** **** ****** ******* issues **** ********* ********** (0 *** *, * for *, ***.) *** failing ** **** *** or **** *********, ********* first ** **** ** the *****.
    • ****** **** ************ **** multiple ***** ** **** vehicle, **** ** ***** one ** ***** ***** being *******.
    • *** ***** ** ***-***** lettering **** ** *****, UPS, ***** *******, ***., only * ***** *** of *** *****.
    • ***-** ******* **** ***** missed ** ******* ****** due ** ***** ****** being *** *** ** gain/digital ***** ****** ***** too ****, **** ********* plate ********** ********* ** fully.

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

    • *** ***********, *** ***** rate ***** ******* *********** by ********* **************, ***** substantially **** *********/******* **** H.264.
    • ********** ** * ****** LPR ****** ***** *********** settings ********* *** **** minimally **** ** ********, but ************* **** ******** in *** *****. ********** ***** cameras ************** ******** ** further ********, ** ** ~25% **** ******** ** scene ** *** ******* (on **** **** **** I7 *.* *** *******).
    • ********* ** ***** **** from * ** ** or ****** ********** ****** as *** ********** ***** filled **** ******, ******* delay ** ***** ******** and ******** ****** ***** at *****.
    • ********* ** *.*** ********** significant ****** ** *** event ******* ** **-** seconds
    • ********** ********** **** **** to ***** *** *** introduced ****** ** ** increase ** ****** ****, nor ********** *****.
    • **** ****** **** ** recorded (** ******) ** retain ***** ******. ******** H.264 ****** *** ** recorded ** ****, *** is *** ********** ** LPR ******.

    *******

    ********* ******** *** ******** three ********, **** *** following ****:

    • ******** *** **** *******: $300 *** **** (******** one ***** *******)
    • ******** *** ******* *******: $1,295 ****
    • ******** *** ***** *******: $645 ****

    *** **** ******* ** required *** ******, **** a ******* ******* ******** for **** ******. * plate ******* **** ** purchased ** ***** **** state ** ******* ** be **** **** *** system. *** *******, ** **** United *******, ******, *** Swiss ******, ***** ********* are ******** (*** ******** in **** **** *** additional).

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

    ******** *** ******** **** accuracy **** ***** ********** are ********, **** ***** integration *** *** ********* Smart ******, ** * price ***** **** ***** than **** ********* *** systems, **** ********* ******, ****** ******* *** advanced ******* ***** ********* and ******* ****** ***** of *******.

    *******, ***** ****** ****** of ********* *********/******* ************ of *****, ***** *** stream **** *** *** processing **** **** ** retained ** ***** ****** are ** ** *****. Also, ****** ** **** PPF ************ ***** ****** preclude *** ******* **** being **** *** ******* surveillance ********.

    *******, ** **** ***** (>40 ***) *********** ** a ****, ********* ********/******** platforms *** ****** * better ******.

    XProtect ****** ***

    ** *** ****** **** be ***** ** **** client ******* ** ****** events. ****** *** ***** in * ******* *** pane, **** ***** ****** directly **** ** **** video. ******** ** ***** switches ** ******** **** in *** ********** ****. An *** *** ** also *****, ******** ***** to **** ******* ****** reads, ****** *** ******** plates (********* ******* *******), and ****** *******.

    **** ***** ******** ** overview ** *** *** in *** ******** ******:

    Server *************

    *** ******** *** ****** runs ** * ******** service, ***** **** *** XProtect ***** *** ********* servers. ********* ********** ** be *** ** * dedicated ******* *** *** the *** ******. ** *** tests, *** *** ****** increased *** **** ** about **% ** ******* (dual ****-**** ***** **, 2.4 ***), **** ****** impact ** ****** *****.

    **** ***** ******* ***** configuration ** *** *** server:

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

    ************ ** *** *** server, ** **** ** XProtect ********* *** ******** issues. ** **** *****, installation ****** ******* *** error ***********. ** ******, installation of ********* *** ****** multiple *****, *** ****** installation *** *** ********, despite ***** ********* ** per *********'* *************.

    ***** ****** *** *********** due ** *** **** and ****** ** *** Milestone ************, ***** *** easily **** **** **** an **** ****** * proper *******. ***** ****** be ******** ** **** ***** time (*****, ********) ****** installation ** ******* *** these ******.

    XProtect *** ****** *************

    ****** ** *** ****** to *** ****** ******** a ***** ************* *******, capturing ** ******* *********** of *** ***** ** view **** ******* ****** in *** *****, *** optional ********** ** * handful ** ******** **** as ********** *********, ****** of ****** ** *** scene, ***.

    *** ***** ******* ** ********** in *** ***** *****:

    ** **** ** ***** video ** ***** ******** parameters **** ** ********** frame ****, ****** ** plates ** *****, ********** time, *** ****:

    Milestone ****** ***************

    ********* ******** ********** *** following ******** *** *** cameras:

    • * ***
    • *****, ******* >**
    • ********** "*** *** ****"

    ***** ******** *** ********, to *** *** *****, since *** **** ******** of ** ******* ********* use *.***, **** ***** used **** ** ******* cases, **** ************. 

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

    *.*** *** ** **** for *** *******. *******, it ********** *********** ********** delay, **** *** **** between *** *** ******* and *** ***** ***** processed ********* **** second (***** *****) ** 10-15 *******.

    **** ***** ******** ** recognition events ******** ******* *** retrieval, ** *** ***** was ********* **** ***** the ****** ***** ********. ******* of **** *****, ******** on ** *** ***** retrieves ***** ***** *** car *** ******* ******, seen *****:

    ********* 

    ********* ********** ***** * FPS *** *** **********. Note **** **** ********* is ******** **** **** and ********* *********, ***** may ** *** ********** or *** ***** *******.

    ********** ********* **** *** recommended * *** *** processing ********** *********** ********** delay, ******* ** ***** H.264, **** *****, *** *** not ******* *********** ***********.

    **********

    ********* **** *** **** specific ****** ** ********** the ****** ** ******* of **********. *******, **** give *** ********** ** 130 *** ******* *** US ******, *** *** PPF *** ****-***** ******.

    ***** ******* *** ******** based ** * ******* recommendation ** *.* ****** per ********* "******" (**** making ** * *********), which ****** ** ***** size *** *****, *** equate ******* ** *** 130/180 *** ******* *****.

    ** ****** ****** *********** from **** ** ** *** and ***** ** *********** difference ** ********** **** or *****.

    Most ****** *********** ******

    ***** *** *** **** common ****** ***** ** our *******:

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

    *** **** ****** ***** we ***** ** *********** *** mistaken **********, **** ** B ***** ******** *** 8 ** * ******** for *. 

    ************, ********* ********** **** simply *** ****, **** commonly *** ***** ** last ********* ** * license *****:

    *** *******: ******** ***** Reads

    ******** *** ********** **** the **** ***** ******** times. ** *** *****, one ** ***** ***** was *******, ****** ****** may ** *********, ** seen *****:

    ***-** *******: ***** *** Gain ******

    *** *** ***** ***** caused ********* **** *********** ***** in ***-** ******* *** high **** ******. ** ***** to ******* ****** ** low *****, ***** ******* of **** *** *******, which ********* ******* *****, partially ** ***** ********* the ******* ** *** license *****. **** ****** complete ****** *** ******** in ***** ***** *** camera *** *******.

    ** **** (*** *** all, **** ** *** Bosch ***-***** **** *****) cameras, **** *** ** manually *******, *** **** darkens ******, *********** ****** them ******** ******* ****** visible ** ** ************.

    Versions *** ******** ****

    ********* ******** ********* **** 9.0c *** **** **** XProtect *** *.**. ******* module **** *** **, with ** ******** ***** modules ****. **** *** recognition ********** *** ***** from **** *** ***** collected **** ******** **** of *******.

    ****** ******** ******** **** as *******:

Comments (13)

We find its hit and miss. Complex and often lacks clear points in which to test and confirm the configuration. We have notice issues where we have had to remove and replace it for another solution after nine months of issues, even the "nothing is wrong its your hardware" only to find a new release fixed other issues - furthermore we have had been massive delays reading the plate up to a minute before it going in a database resulting in the gate opening too late or the system disregarding it. Milestone support is terrible. They do not understand the product enough, the logs files are poor leading to issues tracing any problems even for them.

BTW Milestone tell you that Express supports LPR, you pay your money, install it and then find out from support no it doesn't.. Their website tells a different story.

Currently we have another site which is taking Milestone over 2 days just to get the license to work!

Sorry but i'm on the fence on Milestone LPR, my gut tells me the dacollen system is too processor heavy and too poorly integrated.

Express can not support LPR, as this platform doesn't use a 'Management Server' in it's configuration - only Corporate, Expert and Enterprise use this function to 'centrally control' numerous recording servers.

In lower levels (like Express), they use a separate 'Management Application' to control each individual recording server.

They named each of these interfaces so similarly just to annoy people I think...

Can you confirm the model of processor used? We're trying to gauge the level of performance vs new i7 processors and also their Xeon equivalents.

If you can just get a generic US library, is there any advantage to paying for libraries for individual US states?

Will the generic US library work on Canadian license plates, as they are the same size and shape as US license plates?

The individual libraries are supposed to be "tuned" to specific states, to account for things like differences in character pattern, reflectance, etc. But we haven't seen much difference. We used the NJ library and tested with "Shore to Please" plate, which has stacked characters:

It did not read the stacked characters, and NY and PA plates read just as well as they did using the US generic module.

Because of that, I'd expect Canadian plates to work as well (but I won't guarantee it!), and we'll actually get one to add to the plate repertoire in the future.

it also seems the skew of the license plate is possibly greater than it should be for LPR technology with the above pictures.

The cameras were set up at an angle of incidence of about 19-20 degrees to the road, within Milestone's recommendations (<25 degrees). The vertical angle was very shallow as well, as cameras were only about 7' off the ground.

I am guessing that older analog D1 resolution license plate cameras would not work due to low resolution?

I wouldn't rule them out, but the FOV would need to be about 6' wide at most, and you would need to very carefully set it so you get multiple frames of the plate. Angle of incidence would need to be very shallow.

The up side to SD cameras, if there is one, is that the frame size is smaller, so the server should be able to process higher framerates, instead of the 4 Milestone recommends as their general rule.

Thank you for providing information on setup challenges and the impact on time frames. This is very relevant information.

We were struck with Milestone LPR, Thank you for publishing this article which indeed helped us in geeting out.

Is XProtect LPR 2.0a released as I cant find it in their website and I use 1.0c

Milestone supplied us with an advance (non-beta, but non-public as of test time) copy of 2.0a. They just put out a newsletter last week, however, claiming general availability April 15th, so check back in two days.

H.264 may be used for LPR cameras. However, it introduces significant processing delay, with the time between the car passing and the plate being processed increased from < 1 sec for MJPEG to 10-15 seconds for H.264...

Any thoughts on why decoding a 4 FPS H.264 stream would add 10 seconds of latency? Surely the VMS is capable of decoding multiple concurrent 30FPS H.264 streams in < 1 sec for live view as well as VMD, so what gives?

Makes you think that the application was written for MJPEG initially, and to make it compatible with H.264, they took the easy way out and are just decoding H.264 and re-encoding it as MJPEGs before sending it to the LPR server. Speculation? Justified because their abysmal performance invites it.

Was wondering what the camera processing time stat shows when using h.264. If it's < 1 sec even though taking ten, it's probably not the LPR server using up the time...

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

IPVM Camera Calculator User Manual / Guide on Oct 16, 2019
Learn how to use the IPVM Camera Calculator (updated for Version 3.1). The guide below includes instructions, images, gifs, and videos...
Camera Focusing Tutorial on Oct 14, 2019
Camera focus is fundamental to quality imaging. Mistakes can significantly reduce details, making cameras less effective. In this guide, we...
Verkada Cloud VMS/Cameras Tested on May 02, 2019
Verkada is arguably the most ambitious video surveillance startup in many years. The company is developing their own cameras, their own VMS, their...
PlateSmart LPR Profile on Jan 31, 2019
PlateSmart Technologies claims to "turn any conventional surveillance camera into a license plate recognition camera" We spoke with PlateSmart to...
Milestone Machine Learning Camera Auto-Setting Examined on Mar 13, 2019
Milestone wants to improve image quality using Machine Learning to solve the problem of "a camera doesn't know what it is being used for",...
Axis Live Privacy Shield Analytics Tested on Jun 25, 2019
Privacy is becoming a bigger factor in video surveillance, driven both by increased public awareness and by GDPR. Now, Axis has released Live...
Avigilon ACC7 VMS Tested on Jul 22, 2019
Avigilon's Control Center 7 boldly claims it will "transform live video monitoring" with the new Focus of Attention "AI-enabled" interface. We...
Anyvision Facial Recognition Tested on Aug 21, 2019
Anyvision is aiming for $1 billion in revenue by 2022, backed by $74 million in funding. But does their performance live up to the hype they have...
Camera Calculator V3.1 Release Improves User Experience on Oct 17, 2019
IPVM has released a new version of our Camera Calculator, V3.1, with significant user experience improvements, a new development plan, and an...
Avigilon Facial Recognition 'Appearance Alerts' Tested on Dec 18, 2019
Avigilon has released 'Appearance Alerts' in ACC 7.4, which adds real-time facial recognition. To see how this performed, we tested them with...

Most Recent Industry Reports

IronYun AI Analytics Tested on Feb 17, 2020
Taiwan startup IronYun has raised tens of millions for its "mission to be the leading Artificial Intelligence, big data video software as a service...
Access Control ADA and Disability Laws Tutorial on Feb 17, 2020
Safe access control is paramount, especially for those with disabilities. Most countries have codes to mandate safe building access for those...
ISC West 2020 Removes China Pavilion, No Plans To Cancel Or Postpone on Feb 17, 2020
ISC West plans to go on next month, amidst concerns over coronavirus. However, the Asia / China Pavilion has been removed, show organizers...
Hanwha Wisenet X Plus PTRZ Tested on Feb 14, 2020
Hanwha has released their PTRZ camera, the Wisenet X Plus XNV-6081Z, claiming the "modular design allows for easy installation". We bought and...
IPVM Conference 2020 on Feb 13, 2020
IPVM is excited to announce our 2020 conference. This is the first and only industry event that will be 100% sponsor-free. Like IPVM online, the...
Bosch Dropping Dahua on Feb 13, 2020
Bosch has confirmed to IPVM that it is in the process of dropping Dahua, over the next year, as both IP camera contract manufacturer and recorder...
BluB0X Alleges Lenel, S2, Software House Are Dinosaurs on Feb 13, 2020
BluB0X is running an ad campaign labeling Lenel, S2, Software House, Honeywell, AMAG and more as dinosaurs: In a follow-up email to IPVM,...
London Live Police Face Recognition Visited on Feb 13, 2020
London police have officially begun using live facial recognition in select areas of the UK capital, sparking significant controversy. IPVM...
Converged vs Dedicated Networks For Surveillance Tutorial on Feb 12, 2020
Use the existing network or deploy a new one? This is a critical choice in designing video surveillance systems. Though 'convergence' was a big...