Dahua N3wbie , can’t figure out why IVS events won’t record

In that image I would guess that its simply too dark and not enough contrast, but yes, right idea

Blue box has little or nothing to do with the actual lines. My cameras blue box possible targets well far away from any rule line.

Lets see your image settings, exposure, etc

AND lets see your recorded timeline and recording rules. I dont care about blue boxes, I care about triggers
My camera will only blue-box me near those IVS lines.. away from it, unless i draw another rule out on the sidewalk, it won't blue box me as a target. If i drew an intrusion rule out there it would though...

image settings are self-adaptive and WDR with a setting of 20:


It's bizzare to me that you can see the whole area quite well normally. I think maybe the video I uploaded shows crushed blacks or something as part of the re-encoding just so I could upload it (file size).. but it's very easy to see people in that corner.

Screenshot 2024-07-22 at 1.44.23 PM.pngScreenshot 2024-07-22 at 1.44.12 PM.png
 
The way that you have your rules drawn in the yard and drive are exactly what I was talking about in my example above re needing to think in a 3D space. It's likely that someone walking though a good portion of your lines in the yard won't be picked up until they're well within. You'll probably need to extend the lines up so they cover the road. Yep, I understand that's going to get you alerts that you don't want but that's likely how it's going to work with that view. SMD works because it uses the entire view (and also will get unwanted alerts outside of your rule area).

Do you need to distinguish all of the different zones? If so, that's fine. If not and as a first step even if so, I'd simplify things and get it working well one at a time and work


So this is interesting.. i'm getting the opposite advice from bigredfish that I need to have the lines be more constrained instead of larger/more expansive.. I put them out there on the lawn in such a way as I would imagine people would only start triggering them if they walked onto the lawn toward the camera with their knees or so crossing the line. Is that the wrong way to think about that ?
 
No you are not
 
  • Like
Reactions: JDreaming
No where did I tell you to constrain any lines. I specifically told you to expand them horizontally NOT vertically
 
  • Like
Reactions: JDreaming
So this is interesting.. i'm getting the opposite advice from bigredfish that I need to have the lines be more constrained instead of larger/more expansive.. I put them out there on the lawn in such a way as I would imagine people would only start triggering them if they walked onto the lawn toward the camera with their knees or so crossing the line. Is that the wrong way to think about that ?

And thats correct. But you are leaving a good 10ft of open area between the curb and your lines

Screenshot 2024-07-22 at 10.49.22 AM.png
 
  • Like
Reactions: JDreaming
No, we're just talking about different aspects of the same effect in different ways. And the yard vs porch are different views where that plays a little differently. Also, various ways to get to the same place.
 
  • Like
Reactions: JDreaming
And thats correct. But you are leaving a good 10ft of open area between the curb and your lines

View attachment 199247
Yes, the intent there was to not catch people walking on the sidewalk and only catch people going onto the lawn.

But, I think what you are saying is that in order to do that, I necessarily have to extend it out on the sidewalk or I will not trigger until people are much deeper into the lawn, because not enough of their profile will cross the tripwire or intrusion barrier otherwise .. is that right ?

So, the lesson is that in this case I would extend it out into the sidewalk and then just accept that I would necessarily probably get some sidewalk people as well, and if I really wanted to get everybody on the sidewalk too, I would really have to go into the road..
 
Here, this isn't directly applicable to your particular view but it illustrates the "3D" effect that I was talking about well.

This is a view where I want to watch a boat that I have for any human presence. Looks odd being up in the trees. You might think that you could just draw a box on the ground in the green grass area but that doesn't work well given the perspective of the camera. I have to draw it up higher off of the ground. If you imagine how a person would look when they get close enough to the boat for it to trigger it makes more sense. The gap above where the van is parked avoids picking up people walking by on the sidewalk but once anyone gets close enough to be of interest it will trigger. (There are other tripwires done in a similar way but just clutters it as far as the example.)


Screenshot 2024-07-22 143237.png
 
Except I think thats the opposite of his problem on the porch, but it does show the 3d effect this time with a camera mounted low and looking "up"
 
So I tried to make life simpler to start with a large single IVS perimeter upto the curb as you recommended .. but on testing it seems the camera isn’t picking up hits walking around the yard simply because it’s failing to blue-box the human (me).. despite being in full outdoor lighting and unobscured , with the sensitivity maxed out .. this is with me being fully inside the box too .. there just appears to be no triggering due to no human recognizing.. I went and tried to make sure there was no max/min settings .. and there does not appear to be any boxes drawn when I click on those options .. I even made boxes and deleted them (and saved) to be sure . I’m wondering how much of my issues is due to the human recognition not hitting on me rather than the tripwires /perimeter. I previously had a Reolink cam in this spot and it did a great job of identifying human targets in an identical setup .. so I’m baffled
 
Is this the 180 cam?

What are you recording to? BI? NVR?
 
My camera will only blue-box me near those IVS lines.. away from it, unless i draw another rule out on the sidewalk, it won't blue box me as a target. If i drew an intrusion rule out there it would though...

image settings are self-adaptive and WDR with a setting of 20:


It's bizzare to me that you can see the whole area quite well normally. I think maybe the video I uploaded shows crushed blacks or something as part of the re-encoding just so I could upload it (file size).. but it's very easy to see people in that corner.

View attachment 199242View attachment 199243
Get the camera off of Self-adaptive, and read this: Dahua 4MP IP Camera setting look poor
 
What camera model is that? (possibly some of this could be from the 180 cam, I dont own one)
Are you recording to BI or an NVR?

Do your screens match these relatively speaking?
settingsDay1.jpg settingsDay2.jpg settingsDay3.jpg
 
  • Like
Reactions: JDreaming
What camera model is that? (possibly some of this could be from the 180 cam, I dont own one)
Are you recording to BI or an NVR?

Do your screens match these relatively speaking?
View attachment 199255 View attachment 199256 View attachment 199257


This is the T180 cam. I’m recording to my synology NAS , surveillance station , which seems to have plenty of bandwidth and ram and CPU headroom to spare while recording 9 cameras. It’s all hardwired ethernet of course for the Dahua.

My screens do look similar to these but I’m on self-adaptive instead of day/night , and it does seem to be doing fine on the same set of settings in both cases in terms of how the day /night images look to my eye . A backlight setting WDR of less than 20 leaves not enough brightness /contrast and anything more than 30-40 seems to start washing stuff out and make it look weird . I don’t seem to get significantly better images at night or day by playing with the default brightness/contrast settings.

My exposure setting is on auto , but it doesn’t seem bad ? I’m running 20 FPS at 4096 bitrate, but have tried 15 and 25 and don’t note too much a difference either way in detections.