Because its such a big yard, if your working at gate 1, you need to have gate 2 and 3 armed. otherwise if someone breaks into gate 2 or 3, you'll never know they were there.
Its happened before in the past, so the guys at gate1 hear sirens going off knowing someones walked into gate 2 or 3.
Hey guys,
So I have 30 cameras connected to Blueiris, they are divided into 3 areas of our yard, so 10 cameras in each section.
Each section has an entrance/gate where staff enter and exit, i have setup RFID readers at each gate that pretty much just send an HTTP command to the BI server to...
Bloody Spiders!!!!
Regarding false positives when using CPAI 2.0.8 with Object Detection (YOLOv5 .NET), I have a 85% min confidence filter for "person" and still the bloody spiders are triggering alerts.
is it worth using a custom model to get less false positives? or somehow adjusting the...
I uninstalled the latest version and rolled back to CPAI 2.0.8 and its working fine, im guessing the latest version of CPAI only works with more recent versions of BI, mines almost 1 year out of date.
I decided to upgrade CPAI to the latest version 2.1.10Beta, but before doing so i uninstalled the previous version, deleted all the programdata/programfiles folders, rebooted, installed the latest and let the models download.
Once that had finished, i rebooted, web server starts using default...
Thanks for the tip Actran, yeah i wanted low confidence because further in the background of the image people can sometimes jump the fence and appear small on the screen in the distance.
but i'll give your ratio trick a try...
Thanks
Bloody birds still keep getting detected as a "person", I've tried using the "ipcam-generals" models, still false positives. Tried disabling custom models and just using the standard
YOLO 6.2, but still getting bird people. any ideas guys? i dont really want to increase the min confidence above 60%.
Yup, i know the drill.... lots of the recent updates have fixed something and broken another, but ive been having this greenscreen issue for a while now, just sharing the changes ive made because it finally seemed to resolve the issue for me.
apart from that, the latest recent releases have...
Ok, so after turning off HW accel and cross referencing the camera in question with another camera of the same make/model, i found only one difference.....
The camera coming up green in BI was set to "Consistent" in the cameras webGUI
After switching to Variable, the issue seems to have gone...
Hey guys, I'm still getting this green screen in combination with "no signal" screens.
Weird thing is, its only happening on one of my 15 cameras, and its not always consistent.
I'm Running the very latest version, 5.6.2.4 plus I know its not the cameras issue, the raw stream from the cameras...
Update:
Decided to play around with the raspberry Pi this weekend and make some progress with the "Person Detection System"
So, when cars drive past my place, its all over in about 6 sec before they fall out of range, but people walking by is fine because there is plenty of time to capture...
Thanks dude, yeah pretty sure Kali has an ARM build you can download on their website, but using any flavor of linux will work, just install the aircrack-ng suite
Kali just has all the tools preinstalled.
Yes, having the IMEI number would be great, but thats not broadcasted over wifi frequency range, so all your left with is the metadata that's available to you via the beacon packets.
Thinking back, one of the tricks i used to do to minimize noise, is unscrewing the antenna from the wifi dongle and wrapping the dongle in tinfoil (with small holes)
This limits the range to how far the dongle can sniff beacon packets so your only detecting devices directly around your property.
Yup, i have working code, its archived somewhere on external HDD, it was only about 30 lines of bash script, and im sure someone here who's actually a programmer could do a better job.
is that you?
also, your right, the only requirements for a person to get "spotted" is holding a smartphone...
You cant eliminate the "noise" because the phones (station) MAC is always changing, but the wifi access points SSID's are always the same (unless the user changes them on their router) but that's pretty uncommon.
Hence why i stopped doing this a few years back when apple,android updated their...
Exactly, this is what i used my system for back when MAC's were static and never changed, i could search the MAC address on Wiggle.net, and find an actual house address, wiggle.net works by people who enable monitor mode on there phone and start whats called "war driving", every time they drive...
Anyway, guys alot can be done with this, but since modern versions of iOS and Android make MAC address random now when broadcasting beacon requests, our only use-case for this with cameras would be capturing "timeframes" of when ANY beacon is discovered.
If you capture beacons for say 1 week...
well, you must be one of the special ones ;)
I used to be paranoid too and stitch anti-static bags into "one" of my jean pockets, that way whenever i felt like being tracked, id just swap my phone into the other pocket, it acts like a faraday cage, blocks all signals
I do actually have code for this, it was running at one stage inside a VM, i'll see if i can dig up the bash script from one of my old archive HDDs
but yeah, it would be better written by someone more experience with programing, my script pretty much just did this...
Poll output of airmon...
no one does that theses days, there was a time when people would turn off wifi/bluetooth when not in use to save power, but battery tech is much more optimized now.
Yes in conjunction with cameras, it can get interesting.... for example, before random MACs, if i ever spotted cop cars driving past, i could flag their macs and sent alerts every time they drove past again in the future.
in my neibouhood, somthing interesting is always going down if the cops...