Author Topic: Multiple Cores or Media Director w/Hard Drive for Camera Surveillance Server?  (Read 4585 times)

rocketlynx

  • Veteran
  • ***
  • Posts: 136
    • View Profile
I am running 0810 Alpha2.20 successfully (I know it still has a ways to go, but it's working fine for me).  Since the final install will have approx. (5) Media Directors, three or more extra Orbiters, Insteon or Z-Wave Lighting control (right now it has a CM-11a and X-10), HVAC Thermostat control via Z-Wave, Security System via GE/Caddx NX-8E w/serial interface to Core and approx. (12) Video Cameras, I would like to keep the Camera Processing/Recording off of the Core/Hybrid and on another computer to be recorded and accessed through LinuxMCE. 

I really don't want to tax the Core with the extra duty of having to deal with streaming Video and Audio to MDs, Lighting Scenarios and the monitoring/recording of (12) cameras and I would also like to insure that the video for the cameras is not disturbed by router reloads and the usual LinuxMCE requirements.

So the question is:  What is the best way to accomplish this?


I will be using Blueberry PV cards for Baseband Video Cameras.  and would install a large hard drive in the 2nd. core Or the dedicated Media Director to accomplish this task.

What is the best approach?

Thanks,
Charles
« Last Edit: May 18, 2009, 06:23:19 pm by rocketlynx »

dlewis

  • Guru
  • ****
  • Posts: 401
    • View Profile
Some people have tried to use Zoneminder as a separate server, but I haven't heard anyone successfully integrate Zoneminder with LinuxMCE...

rocketlynx

  • Veteran
  • ***
  • Posts: 136
    • View Profile
Thanks for the quick response.

Some people have tried to use Zoneminder as a separate server, but I haven't heard anyone successfully integrate Zoneminder with LinuxMCE...

I've looked only a little into Zoneminder, but I was hoping to keep this, as much as possible, within LinuxMCE for easy access of the cameras from MDs and insure LMCE compatibility.

Charles

tschak909

  • LinuxMCE God
  • ****
  • Posts: 5549
  • DOES work for LinuxMCE.
    • View Profile
You're thinking too hard, stop it.

LinuxMCE does not record security streams at all, it merely captures single frames from a camera source. It uses sensors for security data, and if there is a security camera bonded to a set of sensors, then the camera will be asked to provide a frame when a security event takes place.

So damn it, just let LMCE handle it, and STOP OVER THINKING IT!

*thwap-you-upside-the-head*

-Thom

rocketlynx

  • Veteran
  • ***
  • Posts: 136
    • View Profile
Thom, My head is hard so I can take the thwap!

You're thinking too hard, stop it.

LinuxMCE does not record security streams at all, it merely captures single frames from a camera source. It uses sensors for security data, and if there is a security camera bonded to a set of sensors, then the camera will be asked to provide a frame when a security event takes place.

OK, got that answer. 

Another question: Upon frame capturing, triggered from a security event associated with a particular camera, is the number of frames captured controllable, or does it just continue until the breech is gone?

Depending on the answer, I'll maybe use this next scenario:

If I need backup of the cameras recorded in time-lapse or in real time, I'll use a recording computer (DVR) with looping 75-ohm non-terminating inputs, and loop each video camera's coax into my bluecherry PV-143's on the core for the scenario that you described above.  Now LinuxMCE and my ease of use on a day to day basis is satisfied, and, If I need a real time backup, I have the second recording computer (DVR) also.  I think I could also use a relay output from LinuxMCE, associated with the particular event sent from the core to the Surveillance DVR's camera input ,to trigger it from time-lapse to real-time for that particular camera.  Maybe a GC-100 or similar approved relay output board.

So damn it, just let LinuxMCE handle it, and STOP OVER THINKING IT!

*thwap-ou-upside-the-head*

-Thom

Like I said, the head is real hard.  Thanks Thom

Charles

« Last Edit: May 18, 2009, 10:00:35 pm by rocketlynx »

tschak909

  • LinuxMCE God
  • ****
  • Posts: 5549
  • DOES work for LinuxMCE.
    • View Profile
When you view a camera, it just sends get video frame commands and recieving frames until you leave the screen.

upon a security breach, it captures a single frame for when the breach is triggered. You could also trigger other commands on the sensor tripped event as well.

-Thom

bulek

  • Administrator
  • wants to work for LinuxMCE
  • *****
  • Posts: 909
  • Living with LMCE
    • View Profile
Hi,

currently you can use Motion under LMCE through motion-wrapper and you can get pretty much all features of motion. So it can record surveillance videos etc..., but videos are still not "seen" on LMCE by default. But you can search in /home/cameras and see that movies are actually there, maybe you can symlink to that directories....

There was an attempt to do similar with Zoneminder too, but didn't end up in any result. I also like Zoneminder as it has better web interface that allows more configuration freedom...

I also remember that Totallymaxed's team did something on integration of Motion enhancement, but I'm not sure what is the current status of that...

HTH,

regards,

Bulek.
Thanks in advance,

regards,

Bulek.

rocketlynx

  • Veteran
  • ***
  • Posts: 136
    • View Profile
Thanks,  My first PV-143 will be delivered tomorrow and I'll begin testing.

Charles