A member emailed us with some concerns and questions about Video Analytic accuracy, which I have summarized below:
I am quite new to NVR setups. I have struggled through a few Hikvision systems. I am inclined to think other brands are not much different. These cameras 2k,3k,4k & 8k with NVRs offer impressive resolution, and the ability to search for events, to easily find, and examine those events. This is useful and convenient.
The notifications and VCA are near totally useless and this is a big disappointment. The line crossing or intrusion boxes are useless because there is no intelligent filtering of the pixel changes. I am finding that even having cloud activity changing the light levels is causing false notifications. There is no way (in my opinion) to make effective adjustments to the NVR to obtain anywhere near an acceptable level of false notifications.
I am assuming it will still be a few years before we see much promise. The article about the performance of HIkvisions Deep In Mind NVR, is a good example of yet more expensive but operationally useless (at this time) technology. So I am wondering if you know of other products available today that may perform better and are in the same (and hopefully less) cost area as the $3,000 Deep In Mind NVR. One client would like reliable notifications for his property where several cameras are affected by roadside vehicle lights (at night). I am guessing virtually all (or perhaps all) consumer rated NVR's (8 camera systems typically) all perform poorly in this regard.
I do not have experience with lower cost NVRs or analytics, so I am hoping we can get some feedback here. Maybe we could get some strategies for configuring his current Hikvision systems.
To add some detail, the referenced DeepInMind article is here: IPVM DeepInMind Tested Terribly
We have also tested other low cost VMD/analytics: Dahua Intrusion Analytics Test and Hikvision VMD and Intrusion Analytics Tested
In general, we have found that these (and other) analytics perform reliably indoors but struggle in the outdoor conditions the member was referencing.