Best way to avoid event segmentation?
Posted: Tue Jan 26, 2010 6:33 pm
I just had an issue where an event being segmented made me lose some very valuable frames. I have a monitor set for modetect mode, and when there is motion for say 60 seconds, instead of recording a 60 second event, it is recording 10 mini events. In this case there was a 12 second gap between two of the events that were critical as the event was part of a theft.
I had noted previously that event segmentation was happening, so I had increased the "Post Event Image Count" from 10 to 15, in hopes that this would keep the event recording during any momentary lack of motion, but I still had what amounts to a huge gap (12 seconds), while someone was apparently moving very slowly?
Please advise on the best way to basically "keep recording for 15 seconds after motion ceases". And will this same setting also have the effect of not ending the event and creating a new one if motion stops for, say 10 seconds?
I am wary of increasing "Pre Event Image Count" because I know that requires more shmem, and I am capturing at 960x720 so the frames are pretty large.
EDIT: Also it looks like this particular monitor is capturing at 10.24 FPS, if that helps any calculations. I just increased my post event image count to 60, which theoretically in my case would record for an additional 5+ seconds. Is this correct? I am still unclear if there is motion that happens within the post event period, if that will extend the event and not create a new one.
I may have to just bite the (disk space) bullet and put it on full mocord mode
I had noted previously that event segmentation was happening, so I had increased the "Post Event Image Count" from 10 to 15, in hopes that this would keep the event recording during any momentary lack of motion, but I still had what amounts to a huge gap (12 seconds), while someone was apparently moving very slowly?
Please advise on the best way to basically "keep recording for 15 seconds after motion ceases". And will this same setting also have the effect of not ending the event and creating a new one if motion stops for, say 10 seconds?
I am wary of increasing "Pre Event Image Count" because I know that requires more shmem, and I am capturing at 960x720 so the frames are pretty large.
EDIT: Also it looks like this particular monitor is capturing at 10.24 FPS, if that helps any calculations. I just increased my post event image count to 60, which theoretically in my case would record for an additional 5+ seconds. Is this correct? I am still unclear if there is motion that happens within the post event period, if that will extend the event and not create a new one.
I may have to just bite the (disk space) bullet and put it on full mocord mode