Best practices for using Object Analytics

Tellu delivers event-based supervision using cameras from Axis. The detection of humans and movements is done using Object Analytics which is a software running on the camera (on-edge).

This is a collection of documents describing the best practices for using Object Analytics, to solve your use case reliably.

Read more about Object Analytics in the website from Axis here: Object Analytics

Contents

Camera placement

It is important that the camera is placed for optimal performance; this applies to distance to the object and camera angle to the area of interest.

Room layout is important

A disadvantageous camera angle might prevent a scenario to trigger, or the camera might have a poor person detection rate. This document describes how the room layout affects the event-based supervision performance.

Defining a scenario

Object Analytics uses custom defined areas and lines in video frames, called scenarios. This document describes best practices on how to set up and define scenarios for different use cases.

Testing the scenario

A scenario should be tested to verify that it works as intended. This document describes how scenarios should be tested.

Using the camera at night

Due to loss of colors in complete darkness at night, the ability to detect a person is slightly reduced. This document describes how to take this into account when using event-based supervision at night time.