What’s new
Upgrade Tesseract4Android to 4.8.0 (Tesseract 5.5.0)
Quick answer
Optical character recognition (OCR) functionality based on Tesseract via Intents
This app bundles OCR functionality (based on Tesseract) that can be called from other apps via Intents. It listens for Intents with action "org.totschnig.ocr.action.RECOGNIZE" and expects an Uri pointing to a JPEG file as data. The recognized text is passed back in the extra "result" as an object of a parcelable data class Text, that must be copied into the client app.
This app is not intended to be useful stand-alone. It was created out of the motivation to not blow up the size of an application that makes use OCR for a feature that is not needed by all users. Currently it is used by My Expenses for scanning receipts.
Before you start using it, you need to configure the language that Tesseract should use, which triggers the download of the necessary trained models, hence the need for INTERNET permission.
How to install
- Download the APK from the button at the top of this page.
- Allow “Install unknown apps” for your browser or file manager only, not for every app.
- Open the downloaded file and confirm the installation.
- If installation fails with “App not installed,” remove an earlier build of the same app first. A signature conflict is the usual cause.
Data safety
- Every file is checked against its published SHA-256 before it is listed.
- Files are scanned for malware on every update.
- The permissions each app requests are listed on its own page.
- Read the full safety policy for how we test, flag, and delist files.
Frequently asked questions
What is OCR (Tesseract)?
Optical character recognition (OCR) functionality based on Tesseract via Intents
Who develops OCR (Tesseract)?
OCR (Tesseract) is developed by Michael Totschnig.
Which Android version does OCR (Tesseract) require?
OCR (Tesseract) requires Android API 21 or later.
What is the current version of OCR (Tesseract)?
The synchronized F-Droid version is 2.5.0.
Which license does OCR (Tesseract) use?
The signed F-Droid metadata lists Apache-2.0.
Where is the source code for OCR (Tesseract)?
https://github.com/mtotschnig/OCR
Comments
Recent approved comments · sorted by newest
View all comments