The OCR Image action allows you to extract the text from an image using OCR (Optical Character Recognition) (specifically using either Apple Text Recognition (v11+ and macOS 10.15+) or the Tesseract OCR library).
The source image can be pasted into the action, or read from the System Clipboard, Trigger Clipboard, a Named Clipboard, or a file, or you can get the image from the screen.
When reading from the screen, you can get the image from all screens, or from a specific screen or window or area on the screen.
You can choose what language or script the text is in. Each language will require a download, typically between 1 and 10 MB, though there are a few scripts that are significantly larger.
As with any OCR, the text may contain errors so you should be careful not to assume the result is 100% accurate.
Note that the Tesseract OCR software prefers black-on-white text, and while it will often work with white-on-black text, it will be more likely to fail in those cases.