SAE Visualizer

Starwit Awareness Engine is processing video material and extracts results likge movements of objects or geo-positions of objects. This component shall visualize these results. It is a Spring Boot app, that connects to SAE's central communication bus and sends data to a ReactJS frontend.

Please refer to central SAE repository for an overview of all SAE functions.

What does it do

Visualizer can show moving objects in various modes. Following image show map view, which places all tracked objects on a map.

For many applications observing actual trajectories are helpful. Thus Visualizer can display trajectory sets from one or more sources. Next image show trajectories captured by one camera.

How to use

This application needs a connection to ValKey/Redis bus of a running SAE instance. Connection details can be configured via application.properties or environment variables.

Once started application provides various views on processed data.

How to deploy

Application is deployed using Kubernetes with Helm. See deployment guide for more details. If you want to run application locally you can also use Docker compose script. Please note, that this is not intended to be used in productive scenarios.

Development Documentation

In order to build application, two steps are necessary. First package Javascript frontend like so:

    cd webclient/app
    npm install

Then Spring Boot can be build with this command:

    mvn clean install -P frontend

Run application locally with:

    java -jar application/target/application-xxx.jar

You can then access application under: http://localhost:8080/sae-visualizer

Configuration

Application is configured via file application.properties. If you create this file in root folder and run app from there, this will override packaged application.properties.

The following configuration will be used to connect to a running Valkey/Redis instance.

spring.redis.host=localhost
spring.redis.port=6379
#spring.redis.password=yourpassword
spring.redis.maxStreamAge=1s

• spring.redis.maxStreamAge is used to decide whether a stream in Valkey is active or not (and should be exposed to the frontend) based on the age of the last message in the stream

Authentication

This application can be protected by an external authentication with Keycloak. In order to use authentication, you need a properly configured Keycloak instance and the following parameters:

#spring.profiles.active=dev # activate for no auth

# Authentication
spring.security.oauth2.client.provider.keycloak.issuer-uri=http://auth.markus-workstation-starwit/realms/aicockpit
spring.security.oauth2.client.registration.keycloak.client-id=aicockpit
spring.security.oauth2.client.registration.keycloak.client-secret=aicockpit
spring.security.oauth2.client.registration.keycloak.scope=openid

License

Software in this repository is licensed under the AGPL-3.0 license. See license agreement for more details.