OCR options: reliably recognize skewed scans, photos and multilingual documents
Good text recognition starts with a clean image. That is why Text-R prepares every page before the Tesseract engine reads it. The software can straighten skewed scans automatically (deskew), detects pages rotated by 90, 180 or 270 degrees and turns them the right way up. A noise filter removes image noise and scanner artifacts, and disturbing table or separator lines can be detected and removed so that they do not interfere with character recognition. This makes text recognition much more successful with smartphone photos of paper pages, faxes and low-contrast originals.
The layout mode lets you decide how a page is read: automatic, as a single column, as a single block of text or as sparse text, for example with forms and business cards. You can select several recognition languages at the same time. Optionally, Text-R also reads barcodes and QR codes and includes them in the result.

