OCR for FireMonkey

Optical Character Recognition component for FireMonkey

OCR for FireMonkey
OCR for FireMonkey demo
Use OCR component to retrieve text from image, for example from scanned paper document.
  • uses Tesseract OCR engine and Leptonica image processing library
  • supports Windows, macOS, Linux, iOS and Android
  • available for Delphi/C++ Builder XE2 - 11.2
  • source code included in registered version
  • royalty free distribution in applications

Download and order


How can I solve "Cannot initialize Tesseract library" error?
Set FOcr.DataPath property to the folder containing Tesseract language data.

How can I increase OCR speed?
Use Tesseract language data from tessdata_fast repository.

How can I increase OCR accuracy?
Use Tesseract language data from tessdata_best repository.

How can I improve OCR output?
Improving the quality of the output

What files should be deployed to the Android device?
Add these files in Delphi Deployment window:
Local Name Platforms Remote Path
libocr.so [Android] library\lib\armeabi-v7a\
libocr.so [Android64] library\lib\arm64-v8a\
all files from tessdata folder assets\internal\tessdata\

How can I solve "[DCC Error] E2597: ld: file not found: ocr.a"?
Copy file "C:\Program Files (x86)\Winsoft\Optical Character Recognition for FireMonkey\Library\iOSDevice\ocr.a" to the folder containing your application.

Related links