Plugins

Plugins screen of Quantotto Management Console displays tree view of currently available plugins by their type.

In general, many parts of Quantotto Environment Monitoring are open for extension and customization. Users can introduce their own functionality into the system by creating small software packages (referred as Quantotto Plugins) in Python. Plugins page allows installing new plugins.

Quantotto also comes with a set of built-in plugins that leverage the same open architecture. The following areas of Quantotto functionality support plugin approach:

• Face Recognition: post processing of regular face detection to cluster similar faces and create recognized subjects classes.
• Detectors: detector plugins that analyze video, audio and other frames data by applying various techniques, such as image processing, audio signal processing, machine learning, neural nets and more
• Triggers: applying various conditions to streams of classifiers’ results to identify specific situations
• Actions: act on triggered situations by sending notifications, integrating with 3rd party systems or initiating any other activity that is required by user’s scenario.

Tree View

Plugins tree shows built-in and custom plugins. Each type of plugins is represented as a folder. You can click on each plugin to display their properties. Built-in plugins wiull have all the fields as read-only.

Most of plugins have specific parameters that need to be set for proper operation. These parameters are configured wither when new node is created under Scenario Flows (trigger and action plugins) or when you add new Classifier.

List of Plugins

Please refer to Quantotto Plugins../architecture/quantotto_plugins.md) page for a list of built-in plugins and for configuration details.

Installing Custom Plugin

You can install your own Plugin by right-clicking on one of the folders representing Plugin type and selecting “Add” from context menu. New plugin card will appear; fill in the following details:

• Plugin name
• Plugin type (will be pre-filled)
• Plugin URI: language://class.path (for example: py3://acme.plugins.my_plugin.MyPlugin)
• Stateful: Yes / No
• Trained : Yes / No

Click on “Save Changes”.

Plugin will be added to the tree view and new ID will be assigined to it.

In order to install software package, select newly added Plugin. Properties card will show along with installation options in the bottom:

• Check Install / Update plugin package
• Click on “Choose File”
• Browse to ZIP file location (containing signed wheel package)

Click Install.

Upon successful upload and installation, checkmark (V) icon will appear next to Plugin name in the tree view.

Now you can add new Scenario Flows nodes (in case of trigger or action plugin) leveraging this plugin or add new classifiers (in case of detector plugin).

Package Structure

ZIP file should contained signed wheel package of your plugin. Creating signed packages is described here