There is a general problem with use of most surveillance systems, notably signal fatigue. Signal fatigue is when the recipient of the signal is no longer able to perceive the signal. This effect is observed in many environment, e.g. nurses in the intensive care unit (aka ICU) of a hospital. A more prosaic example is the use of closed-circuit televsion (aka CCTV) cameras and the operators who watch many screens simultaneously (apparently). This problem is typically solved through signals intelligence; people, process, and automated systems collecting, analyzing, and acting to separate the signal from the noise.

In other words, what signals are most relevent, timely, accurate, or otherwise prioritized versus all the other signals.

The objective function of signals intelligence is two-fold:

1. Situational awareness -- providing high recency, low latency, information for human-in-the-loop processing
2. Automated actions -- taking an action, e.g. turning on light, without human intervention

This project seeks to provide both functions for signals collected from a wide-variety of sensors, especially those provided by Internet-of-things (IoT) sensors available in the consumer marketplace and available as do-it-yourself (DIY) projects in the open-source-software (OSS) community.

Visual object detection and classification

Process signals from cameras to identify person, animal, and vehicle and provide longitudinal analysis and identify opportunities for human or automated intervention.


This repository and the following software:

Further information on Open Horizon


motion addon for Home Assistant

yolo4motion service for Open Horizon


person and car detection

person entry/exit detection


More Information

If you've read all the way to here and you still want more information, you can find me at the following places:

LinkedIn GitHub Twitter