Skip to main content

Plate Recognizer USDOT OCR

The USDOT Docker image runs the OCR engine on-premises without an internet connection, letting you read information from a truck USDOT picture.

To use it, get a USDOT OCR License (select the On-premises variant). An internet connection is required during installation.

Install​

  1. Install Docker on your machine. See the system requirements.

  2. Install USDOT OCR using the Manual steps, which are quicker if you are familiar with Docker.

  3. Call the prediction API

Upgrade​

  1. Stop the running docker containers using the image. Run docker ps -q --filter ancestor=platerecognizer/usdot to get the docker container id. And then docker stop CONTAINER_ID to stop it.
  2. Update the image: docker pull platerecognizer/usdot.
  3. Run the container (see last step of Installation).

Uninstall​

Each license can only be used on a single machine. If you need to move your installation to a new machine, uninstall it from the initial machine first.

  1. On the machine running USDOT OCR On-premises, verify that the Docker container is running. If it is not running, start it.

    • Open a browser and go to http://localhost:8001/. You should see the USDOT OCR welcome message.

  2. Update the restart policy of the container.

    • Find the CONTAINER_ID of the USDOT OCR On-premises container with docker ps.
    • docker update --restart no CONTAINER_ID

  3. Uninstall the license. Replace HOST-IP with the IP address of the machine running the Docker container (localhost for the same machine).

    • curl http://HOST-IP:8001/uninstall/

    or directly in the browser:

    • http://HOST-IP:8001/uninstall/

System Requirements​

For best performance, we recommend running the container from a Linux machine. With a recent processor (7-8 gen Intel CPU with 4 cores), inference should take around 100ms.

Minimum requirements:

  • Use an OS that supports Docker. See the installation guide for how to install on Windows, Mac or Linux.
  • Intel CPU 3rd gen or higher (must support AVX instructions).
  • At least 500MB of free RAM.
  • If you are using VMware, you should disable EVC (Enhanced vMotion Compatibility) to ensure AVX instructions are available to the container.
tip

If you are behind a firewall, whitelist the required IP addresses.