Sometimes there is no way to indicate an application with the supplied automation providers. In this case, OCR activities provide an alternative approach.
Sometimes it is not possible to indicate an application with the provided automation providers. In this case, OCR activities provide an alternative approach.
OCR stands for Optical Character Recognition and provides the ability to read text from images without directly accessing the text field. OCR is not only used to read text from an image, but it can also be used to search for specific text in an image and click or hover over it. In addition, it is possible to wait for specific text to appear or disappear with OCR.
| Activity | Description |
|---|---|
| Get Text by Anchor | Get the text of a defined area related to an anchor. |
| OCR Click | Click on the specified text. |
| Hover Text | Hover over the specific text. |
| Wait Text Appear | Waits a defined time for a certain text to appear (base of a decision). |
| Wait Text Vanish | Waits a defined time for a certain text to vanish (base of a decision). |
| Get Text From PDF with OCR | Returns the text of a PDF by using optical character recognition |
All OCR activities provide the ability to change the OCR Engine activity. This engine provides and executes the actual OCR methodology.
Hyland RPA provides the Tesseract Engine activity within the OCR activities by default. However, the user is free to choose a different (custom) OCR engine.
The OCR activity itself is responsible for providing the engine with the required image from which to read the specific text and perform the desired action. Below you can see the OCR Click activity with and without the Tesseract Engine activity.
Above you can see the OCR Click activity with and without the Tesseract Engine activity
Anchor Image
Speaking of the provided image, in all OCR activities the user must define an anchor region and the region where the specific text is expected. The following screenshot demonstrates this methodology:
The upper dotted line represents the anchor image to which the OCR activity is directed. It will first search for the HYLAND image. The lower dotted line is the area we want to read out with OCR.