Multi model, YOLOV3 and YOLOV4, full-featured, proactive home and business security system

Hi all, I’m pleased to announce that today I’m releasing/unleashing my new project, “StalkedByTheState”.

StalkedByTheState is a highly effective proactive/preventative home and business security system that installs to a production ready appliance on any of Jetson Nano, Jetson Xavier NX or Jetson Xavier AGX in just a couple of commands.

Full details can be found on the install project on github here:

YouTube will be the main medium for providing information about this project, my channel is here:

StalkedByTheState provides:

  • Security state machine that records triggered video events. Supports time of day/day of week scheduled events. Watchdog timers,
    rate based triggers, synthetic combination events.
  • multi-model CNN scanner that can use multiple AI models within the same image even. You
    can choose to use yolov3 in one part of an image, yolov4 in another, or choose to have to triggered confirmed with both.
    Or either… Classification is done via a network service over a websocket so you can add external models to your appliance
    if you like, or models that are not either YoloV3 or YoloV4
  • Scanner that reads camera frames, classifies them and then matches them for triggers can be dynamically enabled and disabled, either
    at the whole camera level or down to a specific notification level
  • Support for auto-review letsencrypt certificates. Can be setup as a reverse proxy to tunnel access to StalkedByTheState
    and your cameras and other devices over https
  • Elicit a real-world response. Supports out of the box Phidget Interfacekit USB I/O controllers for switching relays or
    reading switches, smoke/carbon monoxide detectors and other wired sensors. Read responses from commonly available remote controls
    such as from Klikaanklikuit in the Netherlands. Or control klikaanklikuit devices.
  • Builtin Certificate manager. GUI controlled generation of X509 certificates and importing and exporting of certificates. Use
    self-signed certificates or real ones. Use https internally for device-device communications
  • Websocket based event subscribe interface
  • Resilient mode of operation that runs the OS in read-only mode with a read-write memory overlay layer and the configuration
    stays read-only till it needs updating. Auto-repair on boot so no hung appliances (Waiting for keyboard input for repair).
  • Runs off SSD

I’m also posting this just one time on each of the nano, NX and Xavier channels as it installs to each of those devices and could be of interest to each of those users, hope that’s ok.

Just in time for Christmas. I’m hoping that this will become the killer app that will motivate you to run AI at the end 24/7.

Kim Hendrikse

1 Like

Just to let you know, this is a very mature project. Most of the development was actually done in 2010 and has been in operation in several sites for over 10 years, but it was closed source. Two years ago I added yolov3 triggers as the main source of detection over sensors. And very recently I added the ability to choose either yolov3 or yolov4 full model even in different parts of the image. It has a huge amount of functionality and one can easily setup a very practical home security with it with a low false positive rate. The basic system is very stable, it has been running for years on end without problems.

I’ve put an enormous amount of work into making the installation as easy as possible. The whole system installs with just two commands. And it has a unique mode of operation that protects the device from damage due to power cuts.

It’s well worth the effort to try it out. based on the amount of discussion so far I think I’m having some problems communicating this message properly because normally home security projects based on yolov3 or yolov4 are very popular :-)

I’m happy to work with anyone closely to optimize it properly for their purpose.

At my home I’m scanning 15 cameras continuously, admittedly I’m using an rtx 2080ti as the backend processor and not the Jetsons for myself but the system would typically detect an intruder within a second and I would typically receive a notification in under two seconds.

In my village there are 4x installations of this system that have been operating for more than a year (mostly nanos, one NX) and a 5th one is on the way for a former policeman, as soon as we can receive a Xavier NX.

In the meantime I’ll keep uploading videos showing how to use different features of the system.

Kim Hendrikse

PS. I’m also looking for a person/business than would like to setup such a system in which we video document all of the configuration details. i.e. how all of the zones are laid out, design decisions on how we configure things. How well it’s detecting people etc. With the idea to publish this on YouTube. This doesn’t have to be on an embedded machine or at least not all of it. We could use a high end GPU card and a desktop for the backend processing.

I don’t want to do this with my current home myself, I may be able to detect intruders very easily but I can’t protect myself from smash and grab very much. In the Netherlands you can’t protect yourself very much at all ever :-(

1 Like