Before and after samples

How to read these samples

Each pair shows one page twice. On the left is the page exactly as it came in the scanned PDF. On the right is the same page in the PDF FixMyScan returned, after one crop for the whole book, clean-up and text recognition. The images are scaled down to fit this page, so fine detail is softer here than in the real files.

What a picture can’t show is the text layer: every right-hand page also has invisible, searchable text under it, so you can search, highlight and look words up on an e-reader.

A travel book: yellowed paper

An old travel book with tinted, uneven paper. The paper turns white, the text turns black, and the wide outer margins are cropped away.

The original page is tinted and slightly uneven in tone, with a wide band of empty margin around the text. After the crop, the text block fills the frame. On a 6-inch Kindle that difference decides whether you can read the page without zooming.

A later page from the same book, processed with the same single crop set once for the whole book.

This is a page from further into the same book. Nobody adjusted it by hand: the crop drawn once at the start applied to it too, and the clean-up matched the first page.

A 1792 book: faint print, old type

A Journey Through Spain, 1792. Faint grey print on yellowed paper becomes sharp black text on white.

Printing from the 1790s is often faint, and the long s (ſ) makes it harder still on a grey background. After clean-up the letters are solid black, which is what e-ink displays best.

Another page from the same 1792 book.

Here is a closer look at the same kind of page, at actual reading size:

Five lines from the 1792 book, close up.

A spelling book: dense columns and small type

An old spelling book with dense word columns and a stained, grey background.

Spelling books pack small type into tight columns. The stained background on the left competes with the letters; on the right the columns stand out on white, and the page is easier to scan by eye.

A later page from the same spelling book.

Notes with Greek and English

A page of notes mixing Greek and English. Both scripts come out sharp, and both are read into the text layer.

Pages that mix scripts are a common reason scanned grammars and classics are hard to search. FixMyScan reads Ancient Greek mixed with Latin text, so words in both scripts end up in the hidden text layer.

What to look for in your own scan

When you judge a result, check three things. First, the crop: the text block should fill the page with a slim, even margin, and no words should be cut off at the edges. Second, the tone: plain text pages should be white with solid black letters, while pictures keep their shading. Third, search: open the PDF, search for a word you can see on a page, and it should be found.

Try it on your own book

These samples are clear scans. Very dark, blurred or low-resolution scans improve less, so the honest test is your own book. How it works walks through the steps.

Questions