Get Started#

  • Time to Complete: 30 minutes

  • Programming Language: Python 3

Prerequisites#

Set up the application#

The following instructions assume Docker engine is correctly set up in the host system. If not, follow the installation guide for docker engine.

  1. Clone the edge-ai-suites repository and change into industrial-edge-insights-vision directory. The directory contains the utility scripts required in the instructions that follow.

    Go to the target directory of your choice and clone the suite. If you want to clone a specific release branch, replace main with the desired tag. To learn more on partial cloning, check the Repository Cloning guide.

    git clone --filter=blob:none --sparse --branch main https://github.com/open-edge-platform/edge-ai-suites.git
cd edge-ai-suites
git sparse-checkout set manufacturing-ai-suite
cd manufacturing-ai-suite/industrial-edge-insights-vision

  2. Set the application-specific environment variable file:

    cp .env_pallet-defect-detection .env

    cp .env_pcb-anomaly-detection .env

  3. Edit the following environment variables in the .env file:

    HOST_IP=<HOST_IP>   # IP address of server where DL Streamer Pipeline Server is running.

MINIO_ACCESS_KEY=   # MinIO service & client access key e.g. intel1234
MINIO_SECRET_KEY=   # MinIO service & client secret key e.g. intel1234

MTX_WEBRTCICESERVERS2_0_USERNAME=<username>  # WebRTC credentials e.g. intel1234
MTX_WEBRTCICESERVERS2_0_PASSWORD=<password>

# application directory
SAMPLE_APP=pallet-defect-detection

    HOST_IP=<HOST_IP>   # IP address of server where DL Streamer Pipeline Server is running.

MINIO_ACCESS_KEY=   # MinIO service & client access key e.g. intel1234
MINIO_SECRET_KEY=   # MinIO service & client secret key e.g. intel1234

MTX_WEBRTCICESERVERS2_0_USERNAME=<username>  # WebRTC credentials e.g. intel1234
MTX_WEBRTCICESERVERS2_0_PASSWORD=<password>

# application directory
SAMPLE_APP=pcb-anomaly-detection

  4. Install the prerequisites. Run with sudo if needed.

    ./setup.sh

    This script sets up the application prerequisites, downloads artifacts, sets executable permissions for scripts, etc. Downloaded resource directories are made available to the application via volume mounting in Docker Compose file automatically.

    Note: For the Pallet Defect Detection application, the setup script downloads a pre-trained detection model by default. If you want to train and use your own custom model, see Generate a Model with Geti™.

Deploy the Application#

  1. Start the Docker application:

    The Docker daemon service should start automatically at boot. If not, you can start it manually:

    sudo systemctl start docker

    Note: If you are running multiple instances of the application, start the services using ./run.sh up instead.

    docker compose up -d

  2. Fetch the list of pipeline loaded available to launch:

    ./sample_list.sh

    This lists the pipeline loaded in DL Streamer Pipeline Server.

    Example Output for Pallet Defect Detection:

    Environment variables loaded from [WORKDIR]/manufacturing-ai-suite/industrial-edge-insights-vision/.env
Running sample app: pallet-defect-detection
Checking status of dlstreamer-pipeline-server...
Server reachable. HTTP Status Code: 200
Loaded pipelines:
[
    ...
    {
        "description": "DL Streamer Pipeline Server pipeline",
        "name": "user_defined_pipelines",
        "parameters": {
        "properties": {
            "detection-properties": {
            "element": {
                "format": "element-properties",
                "name": "detection"
            }
            }
        },
        "type": "object"
        },
        "type": "GStreamer",
        "version": "pallet_defect_detection"
    }
    ...
]

  3. Start the sample application with a pipeline:

    ./sample_start.sh -p pallet_defect_detection

    ./sample_start.sh -p pcb_anomaly_detection

    This command will look for the payload for the pipeline specified in the -p argument above, inside the payload.json file and launch a pipeline instance in DL Streamer Pipeline Server.

    IMPORTANT: Before you run sample_start.sh script, make sure that jq is installed on your system. See the troubleshooting guide for more details.

    Example Output for Pallet Defect Detection:

    Environment variables loaded from [WORKDIR]/manufacturing-ai-suite/industrial-edge-insights-vision/.env
Running sample app: pallet-defect-detection
Checking status of dlstreamer-pipeline-server...
Server reachable. HTTP Status Code: 200
Loading payload from [WORKDIR]/manufacturing-ai-suite/industrial-edge-insights-vision/apps/pallet-defect-detection/payload.json
Payload loaded successfully.
Starting pipeline: pallet_defect_detection
Launching pipeline: pallet_defect_detection
Extracting payload for pipeline: pallet_defect_detection
Found 1 payload(s) for pipeline: pallet_defect_detection
Payload for pipeline 'pallet_defect_detection' {"source":{"uri":"file:///home/pipeline-server/resources/videos/warehouse.avi","type":"uri"},"destination":{"frame":{"type":"webrtc","peer-id":"pdd"}},"parameters":{"detection-properties":{"model":"/home/pipeline-server/resources/models/pallet-defect-detection/model.xml","device":"CPU"}}}
Posting payload to REST server at https://<HOST_IP>/api/pipelines/user_defined_pipelines/pallet_defect_detection
Payload for pipeline 'pallet_defect_detection' posted successfully. Response: "4b36b3ce52ad11f0ad60863f511204e2"

    Note: The pipeline uses the pre-trained model downloaded during setup. To replace it with a custom model trained on your own data using Geti™, follow Generate a Model with Geti™ and replace the model.xml and model.bin files in your resources accordingly.

    Note: This will start the pipeline. To view the inference stream on WebRTC, open a browser and navigate to the application URL below. If you are running multiple instances of the application, provide the NGINX_HTTPS_PORT number in the URL for the application instance, i.e., replace <HOST_IP> with <HOST_IP>:<NGINX_HTTPS_PORT>. If you are running a single instance and using an NGINX_HTTPS_PORT other than the default 443, replace 443 with <NGINX_HTTPS_PORT>.

    https://<HOST_IP>/mediamtx/pdd/

    https://<HOST_IP>/mediamtx/anomaly/

  4. Get the status of running pipeline instance(s):

    ./sample_status.sh

    This command lists the statuses of pipeline instances launched during the lifetime of the sample application.

    Example Output for Pallet Defect Detection:

    Environment variables loaded from [WORKDIR]/manufacturing-ai-suite/industrial-edge-insights-vision/.env
Running sample app: pallet-defect-detection
[
{
    "avg_fps": 30.00446179356829,
    "elapsed_time": 36.927825689315796,
    "id": "4b36b3ce52ad11f0ad60863f511204e2",
    "message": "",
    "start_time": 1750956469.620569,
    "state": "RUNNING"
}
]

  5. Stop pipeline instances.

    ./sample_stop.sh

    This command will stop all instances that are currently in the RUNNING state and return their last status.

    Example Output for Pallet Defect Detection:

    No pipelines specified. Stopping all pipeline instances
Environment variables loaded from [WORKDIR]/manufacturing-ai-suite/industrial-edge-insights-vision/.env
Running sample app: pallet-defect-detection
Checking status of dlstreamer-pipeline-server...
Server reachable. HTTP Status Code: 200
Instance list fetched successfully. HTTP Status Code: 200
Found 1 running pipeline instances.
Stopping pipeline instance with ID: 4b36b3ce52ad11f0ad60863f511204e2
Pipeline instance with ID '4b36b3ce52ad11f0ad60863f511204e2' stopped successfully. Response: {
"avg_fps": 30.002200575353214,
"elapsed_time": 63.72864031791687,
"id": "4b36b3ce52ad11f0ad60863f511204e2",
"message": "",
"start_time": 1750956469.620569,
"state": "RUNNING"
}

    To stop a specific instance, identify it with the --id argument. For example, ./sample_stop.sh --id 4b36b3ce52ad11f0ad60863f511204e2

  6. Stop the Docker application.

    Note: If you are running multiple instances of the application, stop the services using ./run.sh down instead.

    docker compose down -v

    This will bring down the services in the application and remove any volumes.

Further Reading#

Troubleshooting#