Catching moving cars by IVS

Cj500

Getting the hang of it
Jul 5, 2018
91
53
LA
I have tried trip lines zig zagged and intrusion boxes for my LPR, but the IVS only catches the last bit of cars going any speed.

Has anyone had success with a scene like this? LPR is perfect, just prefer to be able to review by IVS events versus all footage to look for cars.

Screenshot_20210728-193338.jpg
 
I'm not sure if you're saying that cars are totally missed by IVS or you wish IVS would catch them sooner. If totally missed, I can say that my experience is pretty good. Only occasionally does a car gets through without triggering the IVS, and when it does, it's usually because the car is too big in the image and the green box it generates is too big to trip the IVS. If you see only the tail end of the cars, I have that problem too. The pre-record buffer doesn't seem to work if you're also recording 24x7. This could be fixed in firmware but alas it hasn't ever been fixed.
 
  • Like
Reactions: bigredfish
@tigerwillow1 Good question. It isn't missing it completely, it just shows the last little bit of the car's bumper as it exits the frame.

I tried a trip line very close to the edge and also tried giving a little space for it to recognize the car passing through the trip line. Speed is between 20 and 45 mph.

@EMPIRETECANDY any way to increase the pre-record buffer so it includes a second or two prior to the IVS tripping so it sees the passing car versus just the last fraction of a second?
 
what is the model No. for this cam Cj?
 
For the IVS to work, the camera needs time to see the object and process it and decide if it matches the criteria. For LPR and tighter zooms for a camera, that can be problematic for some field of views. The car is in the field of view for such a short period of time that it will probably struggle more times than not. Especially at night if all it sees are head/taillights and a plate it will miss most of them.

I am usually a fan of the Dahua AI, but for LPR this is an unique situation that you would be better off with motion detection and a min object size instead of trying the vehicle AI. You will obviously probably get some false triggers, but it beats the alternative of IVS missing a lot of them.
 
  • Like
Reactions: bigredfish
Thanks! I may try regular motion detection tonight because I don't use any alerts, so false positives aren't a huge deal.

For the IVS to work, the camera needs time to see the object and process it and decide if it matches the criteria. For LPR and tighter zooms for a camera, that can be problematic for some field of views. The car is in the field of view for such a short period of time that it will probably struggle more times than not. Especially at night if all it sees are head/taillights and a plate it will miss most of them.

I am usually a fan of the Dahua AI, but for LPR this is an unique situation that you would be better off with motion detection and a min object size instead of trying the vehicle AI. You will obviously probably get some false triggers, but it beats the alternative of IVS missing a lot of them.
 
Thanks! I may try regular motion detection tonight because I don't use any alerts, so false positives aren't a huge deal.
I don't want to discourage you from trying this, just be prepared to get more false positives than you want to deal with. Increasing the pre-record buffer isn't an issue. It just plain doesn't work at all if you're also doing continuous recording.
 
I use both IVS and MD. This way in the timeline if the recording, you will see two different color hash marks one on top of the other. Makes sorting out “real” events from bugs and such much easier.