The Arecont 180-degree cameras work nicely with these units. I have a large site that is deploying this configuration and are very pleased with the design. The only complaint I have is that each Railto unit will support either 4 analog cameras with the A4 units or 4 IP cameras with the I4 units. If you use 1080p cameras, only two can be connected to one Rialto blade. This is per VideoIQ. Also, there are only two audio outputs per unit, so if talk-down is desired, makes it more costly to deploy is a single speaker cannot serve more than one camera. I am waiting to hear back from VideoIQ regaring the flexibility of the audio outputs. Do they follow the Rialto blade or can it be prgrammed to work with any camera on the same network.
Testing Avigilon / VideoIQ Rialto
Published Jan 22, 2014 05:00 AM
***** ***** ** ********* ***** ********* only ** ***** *** *******,*************** ********* ******** ****, ******** ***** ***** ******* to ** ****. ** ****** *** Rialto **** * ******* **** ***** manufacturers ****, *****, ****, *** ******** (*** ******** ******** *******), *** *** **** ** ******* VideoIQ's *** ** ****, *** ****.
** ****** ** *** ******, ** open *****, ******* ** **** ********* protection *********:
*** * ******* ***/********, ********** ******** in ******** ** ***** *******:
** ***** ** ****** *** ********* questions:
- *** *** *******'* *** ** **** stack ** ******* *** ****** **** third ***** *******?
- ***** ************ ********* ****?
- **** *********** **** ***** ******* **** and *****?
- **** ***** *** ************* ********?
**** *** *** *** ******** **** this ****:
******* ********
- ********* ** ***** ******* *** **** when ******** ***** ****** *** ***** of ****, ******* ** ****** *** camera. ******** ****** **** **** *** camera **** ******* ************ ******* **** the ****** ***** ****** ******** ****** in.
- ***** *********** **** **** *** ** our **** ******, *** ****** *********** once *** *******' ******* ******** ****** was ****. ******* ****** ****** *** shadows, *** ** *** **** ****** sources ** ********* *******, ****** ***** triggers.
- ********** ***** ****** *** ****** ******** vehicle ******* *** ****** **** ****** activation ***** (**%+). ********** ****** ****** the *** **** *** * *******.
- ******* *********(******) *** *** *********** **** ***** VideoIQ. ****** ********* ***** *** ** used *** *** **** *******, ********* directionality, **** **** ******** *******. ******* require **** ******** ***** *** **** on **** ***** ** *** ******** to ******** **********.
- *** ** ** ************* ** ********* to ******* ** *** ******, **** one (*******) ***** *** *******.
- ********** *********** *** ********* ****** ** sparse (*********** **********, ****** *** *******, etc.), **** ****** ********** ** ** why ******* ** *** *******.
- ************, ** *** ****** **** ******* not ********* ** **** **********, ******* changing *** ******** ********. ********* *** Rialto ********* ***** ****.
****** ***********
- *******'* *** ** ****, *********, ************ **** ***** ***** ******* connected ** *** ****** ** *** daytime. ** *****, *********** *******, **** most ***** ***** ******* ************* ********** ** ****** *** ***** ***********.
- ***** ***** ***** *******, ***** ********* best, ************* **** ********* *** ***** ***** ***** *******, both *** *** *****.
- ** **** *****, ***** ****** ****** 720p ********** *** ***** *** ******** by ***' ** ****. *******, *** light ********* **** *******, **** ** the **** *****, ***** ***-****, *** Sony ***-***** ************ *** ***** ******* in *** *****.
*******
*** ****** ** *** * ****** price ** ***** $*,*** ***, ***** the ****-** ***** *** ***** $*,***.
******** * *****-***** ******* ***** ****** (**** as *****, ****, ****, ***.) ***** *** ~$100, **** ******* ***** *** ****** more ********** ******** ** * ******** ($3,750 *** ****** + ******* **. $3,800 *** *** ****-***). ***** ***** cost ******* ******** **** *********.
*** ****** ** **** ********* ** a********** ******* ******* ** ******** ** to * ** ******** ** ***. ******** ** *******'* *** ** *******, the ****** * ** ********* ** small ******* ******. ***** ******** * third-party ******* ***** ****** ***** ~$*,***, and ** ****-** ***** *** ~$*,***, the ****** ** **** ********* ***** about *** ********, **** ***** *** about *****.
** ******* **** ******** *******, *** Rialto *** ***** ** **********, ********* on ***** ******* *** *******, ** labor ***** ** ****** *** ******* cameras ***** **** ** **** **********.
***************
*******, ***** **** ******* ****** ** well **, ** ****** **** *******'* own ** *****. **** *** **** it ********** ********** ** ** ******** systems ***** ********* ** ******* ******* when ********* **** ******* *** **** prohibitive. ************, *** ***** *********** *** prefer ** **** ******* **** ***** preferred ****** ****, *** ****** ** a **** ******.
Physical ********
**** ***** ***** ******* *** ******** construction ** *** ****** **:
Adding *******/***********
** **** *****, ** ****** *** process *** ****** *******, *** ******* a ***** ******** ** *********** ***** and ******* ** ********:
Detection ***********
***** ********
********* ** ****** ********* **** **** the ******* ******* *** ***** ** view, ** *******'* *********** ***** ****** detect *** ******'* ******** ** * person.
**** *** ******* ****** ****** *** camera, ** *** *** **** *******'* confidence ** *** ****** ************** ** much ***** (**% **. **% *****), despite *** ******* ***** ****** ** the ******.
**********/**********
********** ****** ****** *** ****** ****** significant ********** ** ********* ********. ********** of ********, ** ******* ********* ** this *** ****** ******** ****** *** camera.
**********, ****** ***** **** ******, *** not ***** *** **** *****.
******* *****
**** ***** *** **** ***** **** the ******, ******* ***** *** ********** seen ** *******, ***** ** *** yellow ***** ** *** ***** *****. Other ******* *** *** ***** **** issue. **** **** ***** ********** **** did *** ******* ****** ** **** were **** **** *** *****, *** crossing *** ***** *********.
Outdoor ***** ***********
** ****** *** ****** *** ******* in *** ******: ** **** ***** with ********** ** ******** *****, ~***' long, ** **** **** ***** ********, and * ******* ***, ********* ******** entrance, ~***' ****, ** **** **** human *** ******* ***********.
**** *****
**** ******** **** ***** *** *********** distances ***** ** **** ****:
*****, ** ****** ** **** ********, as **** ** **** ***:
****** *********** ******, **** ***' ***** Axis *** **** ** ***' **** the ***** ***-****.
****, ** ****** *** ******* ** night, **** ***** ****** ******** ** 1 *** *** *****:
********* *********** *** ******* ******* ** all *******, **** ***** *** ***** models **** ** *** ***** ***-**** and **** *****. *** ***** ** simply *** **** **** ******** *** low ** ******** *******, ******* **** about * *** ** *** **** FOV ** ~*.** *** ** *** far *** ** *** *****.
******* ***/********
********* *** ***** ** **** ****** view:
**** ***** ** **** **** ***** the *****:
********* ** *** ***** ******* ****** from *** ** *** ****. ********* of ***** ******* *** ********* ****** than **** ** **-***', *** **** scene *** ****** **** *********, **** less ******** ******* *** ******* *** the ********** **** ** *** ** our ***** *****.
*** ******* **** **** ** ******** detect ******** ** *** ****** ** the ******* ***. ***** ****** **** able ** ************* ******* ** *** ****** ****, 450'+ ****.
*******, ** ****** *** ******* ** this ***** ** *****, ** ****. Light ****** **** ****** * - * lux *** ** *** ********'* ******* lighting.
*********** ***** ********* *************. ********* ** vehicles *** ******** **** ****** *** parking ***, *** *** ****. ********* of *** ***** ******* ******* ** 100-250'. ** **** ****, *** ***** cameras **** ** *** **** *****, Bosch ***-****, *** **** ***** ************ other ******* ** ***'+.
Test *********
*** ******* **** ******** ** *** latest ******** *** ******* ********* ***** to *******. ******* **** ************ ** 1/30s ******* *****, **** ***** ******** left *******. ******* ******** ********:
- ******* ****** **: *.*.*.*****
- *************: *.*.*.*****
- ******** ** *.* **: *.*.*.**
- ******** ** *.* **: *.*.*.**
- **** *****: *.**.**
- ***** ***-****: *.**
- ***** ***-****: *.**
- **** ***-*****: *.*.*
- **** ***-*****: *.*.*
******* **** ******* *.*.*.**** *** ****.
Comments (16)
SB
Steve Bukoski
Jan 22, 2014
Luis Carmona
Jan 22, 2014Good report and one I have been looking forward to.
1. Did you try connecting cameras to both the Rialot and another VMS, and see how the streams react? I can see people trying to use thse for the analytics but their own VMS for recording, since this appears to be able to accomodate.
2. Was there a minimum framerate in the spec's the Rialto said it needed? Steve above mentined using them with Arecont 180's, but the 180's only out so many frames per seond across all 4 sensors. Is it enough for the Rialto?
SB
Steve Bukoski
Jan 22, 2014Luis,
I believe 7ips is the max each sensor of the Arecont dome can put out. When I contacted VideoIQ with the design approach, they had mentioned they initially had issues with the Arecont dome, but worked with them to resolve the issue. I was assured at that time, that the unit would accept the frame rate. As of right now, several of the cameras are running with the analytics functioning properly. At the end of the project there will be over twenty 180-degree domes spreadout over large property.
Luis Carmona
Jan 22, 2014SB
Steve Bukoski
Jan 22, 2014No. I am using the VideoIQ software and Cloudview software. I haven't checked to see if ExacQvision has got them working yet. Last time I was on their supported camera page, the Rialto unit was still incompatible. The integration between VideoIQ and other VMS systems is just not as smooth as using the VideoIQ software.
Ethan Ace
Jan 22, 2014Luis, you specifically mean adding the cameras to both the Rialto and a VMS, right? You're not asking if we recorded the output stream of the Rialto to the VMS?
The answer on the first one is that yes, we did it. Most cameras were fine. The Sony VB630 dropped the stream periodically when added to both Exacq and the Rialto. That was about it.
As far as integrating the output of the Rialto to VMSs, that I have not done yet.
Luis Carmona
Jan 23, 2014Ethan, yes, that's what I meant, attaching the Arecont's to the Rialto and another VMS at the same time.
Does the Rialto actually have the ability to re-broadcast the stream from the unit itself?
Ethan Ace
Jan 23, 2014Yes, it does, with bounding boxes and object IDs embedded. The integration matrix is here.
KL
Keefe Lovgren
Jan 23, 2014Detection of humans performed best when the subject crossed the field of view, as VideoIQ's classifiers could better detect the object's movement as a person.
Ethan,
What is the cause of this or what could be done to compensate for the lower classification of objects moving toward the camera? PIRs have the same issue due to coverage patterns but surveillance utilizes pixels (at least that is my assumption for analytics) so wouldn't it see the change and react accordingly? We are just learning about analytics and ordering some units for our office so this article is greatly appreciated and a great piece. Thank you.
Sagy Amit
Jan 24, 2014Keefe,
All analytics are suffering from this issue equally. The problem is that object crossing the FOV have more distinct pixel change to be considered motion.
When object are moving directly towards the camera, the only pixel change visible is the object growth in size, which typicaly requires a significantly more distance to be established as human.
The best solution for this is to mount the camera higher. if the angle of the camera is steeper you will usually be able to get better results on the Y axis of the FOV.
I am assuming that the IPVM team is using tripods for their tests so mounting the camera higher is not an option.
KL
Keefe Lovgren
Jan 25, 2014Sagy,
Thank you for the information. Is there a recommended mounting height to combat the issue?
Sagy Amit
Jan 27, 2014Keefe,
It is more of an angle issue, so the height realy depends on the distance you are looking at.
so, for example, if you are looking at a distance of 150Ft, 15-20Ft high would be sufficient.
You can use the rule of thumb: Height= 10% of the detection distance.
So, for 1000Ft distance, camera would idealy be mounted at 100Ft high.
It is not always realistic to mount that high, but it is the only way to improve detection towards and away from camera.
GC
Greg Cortina
Feb 01, 2014[IPVM Editor Note: Commenter is from FLIR, thermal camera manufacturer.]
Now do the test tied to a thermal and you will see greater detection reliability. The headlight, flashing lights, sunrise and sunset become non-issues.
AA
Ameen Abdul Khader
Feb 14, 2014Hello. I was told that the ArecontVision AV8185DN, which is the 180 degree Panoramic view camera is not supported by Rialto for Video Analytics, so I am surprised to see the notes above! If this is true, how is the connectivity? The camera has one output cable but esentially carries the video from 4 sensors; so using a single channel Rialto would work? The application is more about using the Panoramic camera for Perimeter Surveillance and Traffic Monitoring.
SL
Steve Lessard
Sep 28, 2015Anybody have tried the Rialto with thermal cameras, like Flir FC series?
Brian Karas
Sep 29, 2015Steve - yes, there have been several thousand channels of Rialto deployed with various thermal cameras. The firmware has a "thermal/FLIR" mode specifically optimized for thermal camera input, both in IP and analog formats.