Hikvision 2432 sending trigger to BI - won't stop recording

[erkme73]

[SOLVED] Hikvision 2432 sending trigger to BI - won't stop recording

@fenderman - I read in your post from mid-last year that you were stymied by BI refusing to stop recording after detecting PIR motion. I'm wondering if you ever found a solution.

Here are my observations:

When using an alarm that set sets the "A<-1" bit to active, BI latches and doesn't see that the bit gets changed to inactive.

While I haven't played around with the PIR sensor (which does use the A<-1 bit), I was using an external digital input (a photocell beam that would short the DIO when the beam is broken). It works very well, in that the bit does toggle to active, and then drops to inactive 5s after the beam is restored.

However, BI does not stop recording.

Oddly enough, if I use a trigger that is internal to the camera (i.e. motion detection, line crossing detection, etc), while the A<-1 bit does toggle between active/inactive, BI DOES actually stop recording X seconds (defined in the break-time) as it should.

So it seems the camera may actually be sending a different bit for internal alarms vs. DIO or PIR alarms. And for that to be the case, BI is probably polling a different bit than the one we see when we execute the URL http://192.168.2.XXX/IO/outputs/1/status/.

I'd love to be able to use my photocell beam, but it behaves exactly the way you described the PIR.

 

[fenderman]

Last time i tested it I used a 2432 with 5.1.6 firmware and an older version of bi-i installed that camera for a friend so i no longer have it.....This time im using a laview hikvision clone with 5.2.0 firmware...i followed the instructions and it works just fine...both for pir and for motion...not sure if its the firmware or something else...
Edit: just looked at it again, and it seems it started to continuously record and not stop...gonna mess with it..

 

[erkme73]

Thanks. The 2432 I'm using is V5.2.0 build 140721.

I'm just having a hard time reconciling how a single toggle between in/active can be treated differently by BI. There must be some other variable that BI is looking at.

 

[erkme73]

Alright, I just figured out what the problem is (at least on my system).

It doesn't matter what is tripping the flag (internal motion, line-crossing, intrusion, or my optical sensor on the DIO of the camera) - but if you keep the Motion/Trigger tab in BI enabled, the recording will continue indefinitely.

At first I thought it was only the external DIO trigger that was doing it, but as it turns out, it's a combination of having the "trigger using camera's digital input or motion detection" option tick PLUS the BI motion detection being enabled.

So, not only are they mutually exclusive, there is some sort of a bug that traps the recording, causing it to become continuous.

 

[fenderman]

Nice catch!

 

[erkme73]

I forwarded my findings to Ken last night. This afternoon, he wrote back saying he found the bug - the pathway for the DIO check is through the PTZ control thread - which stop the motion detection. He said it will be fixed with the next release. Hopefully it'll address the continuous record if both DIO and MD are enabled.

 

[fenderman]

Awesome! Thanks.

 

[erkme73]

With the release of .14, the problem of a dual triggered (DIO input and BI MD) has definitely been resolved. The camera will now record on BI MD and after the "break time" has elapsed, will return to idle waiting for the next trigger. Same thing happens when it's triggered by the camera (either DIO, camera-based motion or line-crossing, etc). But, and this is a pretty big but... If BI MD triggers a recording, then while it's recording, BI ignores additional DIO inputs.

This is significant in my case, since only the DIO trigger sends me a push alert. Thus, if someone approaches my front door, and triggers the BI motion detect BEFORE they cross the line that triggers the camera's alert bit, then I'll never get a notification.

Obviously this is a huge step in the right direction, but for it to be considered useful (for me), the two triggers must fire completely independently of each other.

I've sent another email to Ken to clarify if I'm just not expecting too much for a single camera channel on BI to be this selective in how it reacts to two different scenarios occurring in quick succession.

 

[ruppmeister]

I, like fenderman, am using the laview cube clone and wanting to know there has been any resolution to this problem as of yet? @erkme73, I have the same problem that I am triggering the alerts to mobile with DIO and not getting the event triggers when BI motion is on in addition to camera triggers. I end up having to clone the camera in BI to set one for MD and the other for PIR from camera. I wish we only needed a single camera instance as I have limited CPU resources at this time.

Thanks for any updates you both have.

 

[fenderman]

I, like fenderman, am using the laview cube clone and wanting to know there has been any resolution to this problem as of yet? @erkme73, I have the same problem that I am triggering the alerts to mobile with DIO and not getting the event triggers when BI motion is on in addition to camera triggers. I end up having to clone the camera in BI to set one for MD and the other for PIR from camera. I wish we only needed a single camera instance as I have limited CPU resources at this time.

Thanks for any updates you both have.

Cloning the camera should not have any significant effect on cpu consumption...

 

[ruppmeister]

I must have done something incorrect when I set it up then because I took a pretty good hit when I did it. I even turned off the record capability and hid it out of view but that didn't help much. I'll look over my settings once more to make sure it is setup correctly.

 

[erkme73]

I concur with fenderman... I don't see any difference in CPU - but what does bother me is that you can't hide the duplicate images from the webserver or app. The console will hide it fine, but given that my BI server doesn't even have a physical monitor connected to it, it doesn't do me any good.

I've asked Ken on several occasions, and he seemed interested, but each time there was some rash of fires going on that preempted the improvement.

 

[erkme73]

@ruppmeister - perhaps send Ken an email as well. Maybe if he's getting it from all directions, he'll re-prioritize. I just hate being the only one to keep bringing it up. I feel like I'm annoying him...

 

[ruppmeister]

I just got back to my BI Server and tried again to add the camera just for the purpose of triggering from the DIO. Turns out I must have done something wrong the first time as I didn't see any noticeable difference in the processor usage after adding it this time. Thanks for the feedback @fenderman and @erkme73

Now if we could just get the event triggers from Hik cams without the PIR in BI... THAT would be great (line crossing anyone?).