> ## Documentation Index
> Fetch the complete documentation index at: https://dragonwingdocs.qualcomm.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Run basic OCR with ocr_service

> The ocr_service sample application enables a service that provides the optical character recognition (OCR) function. It captures the image topic from the ROS system and publishes the result with ocr_topic.

## Pipeline flow for `ocr_service`

The following figure shows the pipeline flow among the relevant client/server and ROS nodes, with the basic messages and data transfer channels.

<Frame caption="The ocr_service pipeline.">
  <img src="https://mintcdn.com/qualcomm-prod/eHVY1H1lLEVhgq58/SDKs/QIR-SDK-Ubuntu/images/image7.png?fit=max&auto=format&n=eHVY1H1lLEVhgq58&q=85&s=4a24e8966fd62084f6361684ff21776a" style={{width: "70%"}} alt="OCR_Client calling OCR_Server over ocr_msg.srv, the server forking OCR_Node, which subscribes to image_topic and publishes OCR_Topic." width="2608" height="2252" data-path="SDKs/QIR-SDK-Ubuntu/images/image7.png" />
</Frame>

## ROS nodes used in the `ocr_service` pipeline

The following table lists the ROS nodes used in the `ocr_service` pipeline and describes what each node does.

| ROS node      | Description                                                                                       |
| ------------- | ------------------------------------------------------------------------------------------------- |
| `OCR_CLIENT`  | Sends the optical character recognition (OCR) request to `ocr_server` and gets the server result. |
| `OCR_SERVER`  | Gets the OCR request from `ocr_server` , and forks the `ocr_rosnode` according to the request.    |
| `OCR_ROSNODE` | Subscribes to the image topic according to the request.                                           |

## ROS topics used in the `ocr_service` pipeline

The following table lists the ROS topics used in the `ocr_service` pipeline, with the message type and the publisher of each topic.

| ROS topic               | Type                        | Published by       |
| ----------------------- | --------------------------- | ------------------ |
| `/image_topic_name`     | `< sensor_msgs.msg.Image >` | Any ROS node       |
| `/ocr_image_topic_name` | `< std_msgs/msg/String >`   | `ocr-rosnode` node |

## Prerequisites

You have set up the device, installed ROS2 Jazzy and the Qualcomm Intelligent Robotics (QIR) SDK on the device according to [Install the QIR SDK](./install-the-qir-sdk).

## Run out-of-the-box `ocr_service`

Run `ocr_service` with the following commands:

```bash Run ocr_service theme={null}
source /opt/ros/jazzy/setup.bash
ros2 run ocr_service ocr_server
ros2 run ocr_service ocr_testnode --topic <image_topic> --picture <image_path>
ros2 run ocr_service ocr_client <image_topic>
```

### Example

This example shows the commands you can run in the three terminals:

* In terminal 1:

  ```bash Start the OCR server theme={null}
  source /opt/ros/jazzy/setup.bash
  ros2 run ocr_service ocr_server
  ```

* In terminal 2:

  ```bash Publish a test image theme={null}
  source /opt/ros/jazzy/setup.bash
  scp <pic_path>/test.png ubuntu@IP:/home/ubuntu/image
  ros2 run ocr_service ocr_testnode --topic test --picture /home/ubuntu/image/test.png
  ```

* In terminal 3:

  ```bash Run the OCR client theme={null}
  source /opt/ros/jazzy/setup.bash
  ros2 run ocr_service ocr_client test
  ```