

Both recognized Arabic page and section numbers with equal high accuracy. Acrobat failed here it translated "vii" as "Vll" (those are "ells") and "ix" as "IX", for example. * Cisdem recognized the Roman numeral page numbers accurately. * Both apps recognized letters with high accuracy, with Cisdem occasionally missing a space between words, but otherwise matching Acrobat in accuracy.

Cisdem's deskew gave results that were more obviously inhomogeneous (e.g., aligning one side of the page nearly perfectly, but the other side less so). Subjectively, Acrobat's deskew seemed slightly better on all the pages. In deskewing, Acrobat appears to favor vertical alignment and Cidsdem horizontal alignment, e.g., on the title page, Acrobat accurately fixed a full-page vertical line, but was less accurate with a 3/4 page horizontal line. I didn't see deskew mentioned anywhere on the Cisdem web site it's important and they really should mention it (I almost didn't even try it because deskew wasn't mentioned). It was from a math textbook, so not the most common English. I was converting the scanned PDF to a PDF with text. The Cisdem demo will work on up to 3pp, so I processed the first 3 pages of a PDF from a 600 DPI scan of a book, including a title page with vertical and horizontal lines and text, the copyright page with some centered text and a paragraph of normal text, and a page from the TOC. 9) version of Adobe Acrobat Pro (included with my scanner when I bought it several years ago).

Convert encrypted PDF Files thoroughly for the encrypted PDFs, Cisdem PDF Converter OCR could unlock the restriction of user password without data loss and then convert file to the format you prefer.It can let you extract images from large PDF file quickly.And it also can work as a PDF Creator, PDF Password Remover, PDF Compressor to meet all of your PDF processing needs. It will maintain the original layout and file quality with its built-in superior OCR technology. Cisdem PDF Converter OCR for Mac comes to help you convert any PDFs, scanned documents, and images to editable and searchable text and image documents.
