Image Preprocessing (Overview)

Once document pages are loaded, ABBYY products offers a variety of image processing options which prepare document images in a way to deliver the best OCR results:

  • Image cleaning routines to remove noise and garbage
  • Optimisation of images from digital cameras,e.g. straighten curved text lines
  • Dual-page splitting
  • Different algorithms for skew correction up to 20 degrees
  • Built-in adaptive binarisation and texture filtering

Also the improved binarization in V10 Technologies have an impact on overall recognition quality and processing speed.

The level of control is different, depending on the target audience

  • ABBYY “out of the box products” allow the users / admins to select some basic options
  • The toolkits like FineReader Engine have a full API for image pre-processing:


More details, please see FineReader Engine sample:Image Preprocessing for OCR


More details, please see FineReader Engine sample: Camera OCR Code-Sample

Video

Related Articles

  • Mobile Imaging SDK - separate ABBYY SDK to enhance images on a mobile device (no OCR inside)