OCR Recognize File

OCR Recognize File

Type: ocr-recognize  ·  Package: OCR Activities v1.2.5  ·  Output: string

Recognizes and returns all the text of an image file (.png, .jpg, .bmp, .tiff). It is the core activity of the package: read the content of an image or a scanned document and then extract data with expressions.

How it works

Runs the image through the Tesseract engine and returns the recognized text, in the reading order detected by the engine (line breaks between lines and blank lines between blocks). Set the text’s language to improve accuracy, and use preprocess to clean up low-quality images. See language and preprocessing.

Recognition runs on the robot’s machine and takes longer the larger the image is; the original file is not changed.

Required parameters

Parameter	Type	Default	Description
path	string	—	Path to the image file (.png, .jpg, .bmp, .tiff).

Optional parameters

Parameter	Type	Default	Description
language	string	"eng"	Text language: Tesseract code (eng, spa, eng+spa) or regional tag (es-CO, en-US).
preprocess	string	"none"	Prior cleanup: none · auto · deskew · binarize. Since 1.2.0.

Output

Returns a string with the recognized text. If the image has no readable text, it returns an empty or almost empty string (only line breaks); check it with isEmpty(trim(text)) or by validating the data you look for.

Example

Read a scanned invoice in Spanish and extract the total:

OCR Recognize File   path = scanPath   language = "spa"   preprocess = "auto"   → output: text
Set Variable   name = total   value = regexFind(text, "(?<=Total:\s*\$?)[\d.,]+")
Log   message = "Total read: " + total

If the text comes out with errors

Set the right language ("spa" for Spanish) and switch preprocessing to "auto". If it still fails, the problem is usually image quality: you need a higher-resolution scan. Remember to validate what you extract.

Common errors

Message	Cause and fix
File not found.	The path does not exist on the robot’s machine.
'path' is required.	path is missing or the expression returned nothing.
Tesseract language '…' not found at '…'. Available: [eng, spa].	The language code is not installed. Use one of the available ones.
Could not read the image.	With preprocess, the file is not a valid image or is damaged.
OpenCV native init failed: …	A native preprocessing component is missing on the machine. Reinstall the package or use preprocess = "none".

Related activities

• OCR Extract Pattern — extract specific fields with regex in one step.
• OCR Find Text — locate where a text is.
• PDF Get Text — for PDFs with text (no OCR).