Skip to main content

Manual Installation Steps

Follow the steps below to manually install VisionAlert On-premises with Docker.

danger

For the commands below, make sure to replace YOUR_LICENSE_KEY with your License Key and YOUR_API_TOKEN with your API Token. Get your token and license key. Use On-premises Licenses.

List of Docker Images​

The Docker image is available for linux/amd64 architecture. The image structure is:

platerecognizer/vision-alert:<version>

Examples:

  • Latest version: platerecognizer/vision-alert:latest or platerecognizer/vision-alert

Installation​

Abbreviated installation steps are below. For detailed instructions, see Install Docker Engine on Linux and the associated FAQ.

  1. Sign up and log in.

  2. Get a VisionAlert On-premises license from the VisionAlert page or contact us.

  3. Install Docker on your server. See requirements.

  4. Get our On-premises image:

    docker pull platerecognizer/vision-alert

  5. Install and run the Docker container:

    docker run --restart="unless-stopped" \
      -d --name vision-alert \
      -p 8080:8080 \
      -p 2022:2022 \
      -v vision-alert-data:/user-data \
      -e TOKEN=YOUR_API_TOKEN \
      -e LICENSE_KEY=YOUR_LICENSE_KEY \
      -e ADMIN_EMAIL=YOUR_ADMIN_EMAIL \
      -e ADMIN_PASSWORD=YOUR_ADMIN_PASSWORD \
      -e SITE_URL=http://YOUR_SERVER_IP:8080 \
      platerecognizer/vision-alert
    • Replace YOUR_ADMIN_EMAIL and YOUR_ADMIN_PASSWORD with the credentials for the first administrator account.
    • Replace YOUR_SERVER_IP with the IP address or hostname that users and cameras will use to reach the server.
    • See Docker Configuration for all available options.

  6. Check that the container started:

    docker logs -f vision-alert

Open the Dashboard​

Open http://YOUR_SERVER_IP:8080 in a browser and sign in with the ADMIN_EMAIL and ADMIN_PASSWORD you set above. This login is local to your on-premises instance and does not need to match your Plate Recognizer account.

The on-premises instance serves the same endpoints as the VisionAlert API. Replace https://api.platerecognizer.com/v1/ with http://YOUR_SERVER_IP:8080/v1/ in the examples, and create an API token at http://YOUR_SERVER_IP:8080/accounts/tokens/.

Connect Your Cameras​

The on-premises edition currently supports cameras that push snapshots over SFTP to the built-in SFTP server on port 2022. Plain FTP is not available. Direct integrations with Cisco Meraki, Nx Witness, and OpenEye are only available in VisionAlert Cloud.

  1. Point your cameras or VMS at YOUR_SERVER_IP on port 2022 over SFTP. Any camera or VMS that can upload images to an SFTP server is supported.
  2. Upload snapshots using the same folder structure as VisionAlert Cloud: /vision-alert/CAMERA_SERIAL/ or /vision-alert/TAG_NAME/CAMERA_SERIAL/. See the FTP Configuration Guide for details.

Each camera is limited to one uploaded log per collection window. Extra uploads within the window are ignored with the message Camera already uploaded a log recently.

Alerts are delivered through webhooks.