
digiKam's OCR Text Converter produces recognized text from images through Tesseract. Choose whether that result belongs in a separate text file or metadata, then review the characters against the source image. Sharpening the photograph and correcting the OCR transcript are different operations.
The OCR Text Converter manual documents the tool, external engine requirement and result destinations. This is a text-extraction workflow, not an editable reconstruction of the original page layout.
Confirm that the recognition engine is available
Select the scanned images and choose Tools > OCR Text Converter. Inspect the Text recognition tab for the detected Tesseract program/version.
If the program is absent, follow the relevant installation documentation rather than changing image resolution repeatedly to solve a missing-engine problem. This article does not install Tesseract or claim a particular bundled version.
Keep a small representative page available while configuring recognition. A clean printed paragraph and a damaged handwritten note are different input challenges; a successful check on one does not establish accuracy on the other.
Choose language and page interpretation deliberately
Inspect the available language support and segmentation mode. A whole page, a single line and sparse isolated text provide different assumptions to the recognition process.
For a hypothetical bilingual document, verify the required language modules instead of assuming that default Latin-alphabet handling recognizes every script. Do not treat a visually correct-looking character sequence as proof that names or dates are right.
The scanned-PDF text guide explains why photographed words are initially pixels. OCR supplies another representation; it does not turn those original pixels into the same editable text objects used when the page was designed.
Set where the recognized text will be stored
Use Store result in to choose a separate text file or the documented metadata destination. A file in the image's directory and an alternate-language XMP entry have different downstream uses.
For a transcript handoff, confirm that the recipient receives the actual text artifact. For a catalog-search workflow, confirm that the intended metadata entry contains the reviewed text. Neither is established by seeing a successful OCR progress indicator.
Translation is an additional optional step using an online engine. Do not enable it merely to obtain source-language OCR, and do not describe that optional online operation as entirely local recognition.
Review the transcript against the source
Use Text Review to select each processed item, inspect its recognized text and make corrections. Spell checking can help with ordinary words but does not verify a proper name, historical date or product code.
The Live Text editing boundary guide illustrates why editing extracted text does not change words printed in the photograph.
For a proposed acceptance check, compare the first and last lines, a number and an important name against the image. Then open or inspect the chosen output destination. The deliverable is reviewed text associated with its source, not a claim that OCR proved the document's authenticity or reproduced every layout feature.
Related tasks
- Why Can't I Edit the Words in a Scanned PDF Imported into Canva?
- Does Editing Text Copied with Live Text Change the Original Photo?
Sources and scope
- digiKam Manual: OCR Text Converter: Tesseract is an external prerequisite. The output is editable recognized text requiring review, not corrected image pixels.
The official digiKam 9.2.0 manual was opened on October 23, 2026 (Asia/Shanghai). These are documentation-based instructions with proposed examples and acceptance checks, not hands-on test results. Check controls and file associations against the installed application version.
Prepared by Imagild Editorial, the company editorial team responsible for these guides. AI assisted research and drafting. This is documentation-based guidance with proposed examples, not a claim of hands-on testing or an independent product review. Report a correction.
Cover: Mak Cézar / Pexels, Pexels License. This is an illustrative source photograph, not an application screenshot or a generated test result.