False Positive Motion Detection Due to Lighting

Have some questions or having issues with your IP Camera(s), Post them here for the mods and other users to assist you with.
Post Reply
zkaps
Posts: 19
Joined: Mon Dec 13, 2021 6:23 pm

False Positive Motion Detection Due to Lighting

Post by zkaps »

Quick question on reducing false positive motion detection due to quick changes in lighting conditions in the detection area for camera IP2M-841 (e.g., cloudy day and the sun pops out).

Of the two parameters Threshold and Sensitivity, which one has the most impact for the above situation ? No other motion in the detection area.

Thanks,
Zkaps
User avatar
Revo2Maxx
Site Admin
Posts: 5805
Joined: Sat Jun 15, 2019 3:05 pm

Re: False Positive Motion Detection Due to Lighting

Post by Revo2Maxx »

Higher Threshold if I remember right, There was a time I tried to mess with this. However seeing I live in an area with High Humidity, rain, snow and bugs I have since gave up and just leave it where it was when installed.

However Threshold the higher the number the more area in the scene needs to have motion but when it comes to shadow and light that covers the whole grid of the area fast and most times all at once that changes make it so someone walking up to the camera would not set it off anymore in the end as well..

Sensitivity number smaller and threshold number higher, make the adjustments little at a time and see if any small adjustment helps. Normally it won't. Changing the cameras area of light in lens is better in most cases
Here to help the best I can.
Be Safe
zkaps
Posts: 19
Joined: Mon Dec 13, 2021 6:23 pm

Re: False Positive Motion Detection Due to Lighting

Post by zkaps »

I tried several Sensitivity/Threshold combinations so I think I am at a point where I have a much better understanding of the camera's capabilities (IP2M-841) and how to manage them.

A Sensitivity of 60 and Threshold of 5 seems to be a reasonable combination. I had tried both values in the past but never together. This combination has reduced false positives except in situations where the lighting condition changes drastically such as when Christmas lighting goes on or off; this is an exception, cannot do anything about this.

In the testing that I did over the past few days, I observed that the Threshold value is a major contributor to the camera's "D" in the DORI standard. With the Sensitivity setting kept at 60 for all testing scenarios, values above 5 for Threshold start reducing the camera's motion detection range drastically. With a value of 5, the camera consistently detects motion at 35+ ft. With a value of 10, the camera consistently misses motion above about 18 ft. Moving forward if false positives become an issue, my approach will be adjusting Sensitivity and leaving Threshold at 5.

Zkaps
davidt
Posts: 5
Joined: Fri Nov 05, 2021 8:58 pm

Re: False Positive Motion Detection Due to Lighting

Post by davidt »

I have major issues with Bug and Moths setting the camera off like an animal :(

The one thing I miss with my Arlo camera's was the heuristics, it could determine between animals (Dogs, cats, birds etc) and humans, vehicles and larger objects.

It would seem that the Amcrest Camera's have poorly programmed heuristics, on the mobile app you can turn on human detection, but not on the web application, none of that makes sense and maybe why when you turn it on in the app it doesn't work.

Right now I have sensitivities set to 50 and threshold to 8, I had the threshold set to 20 and reduced false alarms drastically, but the camera stopped triggering in regions where I had asked it too, so the above comment about it reducing its field of view/coverage is accurate, and that is such a sad thing because it tells me that the programming of the firmware is bad.

I love the hardware and think they are well made, but the software lets these things down quite drastically, looks like I have to look at other vendors again, I was hoping that Amcrest would be the primary vendor for me, but sadly this wont be the case :(
Post Reply