ZoneMinder Forums

Hi.

First of all I'm sorry of my poor English.

I'm doing a surveillance ZoneMinder 1.22.3 based system with 4 AXIS 210 and 5 AXIS 212 PTZ.

What I know is that ZoneMinder can detect the motion of a camera (with the mode Modect) monitoring and analysing continously the frames (with the obviously microprocessor consumption) and doing nothing, unless with external events (with the mode Nodect).

A single CPU handling 7 Cameras it's so hard for a standard PC, so externalyzing the analysing frames processes would be fine. Axis cameras have web servers that monitorise and analyse the frames catched by the camera, so, do you know if I could monitorise and analyse the frames on the camera for when a movement is detected send a event over the local network to be catched by ZoneMinder and, so then, generate for example a MPEG video ON ZONEMINDER (when a event were reported, ZoneMinder would get the control)?

¿Do you know if it is possible? Maybe via PERL?

P.D: The cameras and the PC with ubuntu and ZoneMinder installed would be on the same SubNet.

Thanks for your time

As far as I know this is still not possible as there is no way to get the status information for the analysis process in any of the IP cameras. This has been discussed before and is sort of a wish list item, but there is nothing to be done till the camera manufactures provide a means to do so.

I've seen on Axis cameras functionalities, that it has a API to customize the embebbed webserver. I've also seen that it's possible to add TCP notifications when a configured event happens (obviously, configured on the camera webserver, not in Zoneminder), so then, it may be possible to have a "nodect" monitor and, with a TCP notification (catched maybe by another webserver in the same machine than Zoneminder, or maybe by a simple program in C) say to the Zoneminder webserver: "Ey! Change monitor X from nodect, to modect" (the communication between the Zoneminder webserver and the TCP Notifications catcher webserver it may be via PERL i guess).

I'm case I'm having a CPU handling over 11 cameras, and saving job to the processor it would be great to perform another activities, or maybe increase the functionalities, and the CPU doesn't suffer so much.

I thing that doing that is great, because you can use the camera processing capabilities, and it's best a distributed job than a single CPU doing all the job.

What do you think about that? Any ideas? It may be interesting to hear our opinions.

my understanding is that it is to slow and unreliable to be able to implement and that there are no current plans to do so.