OCR

Why OCR matters

A scanned page looks like text to you, but to a computer it is a flat image. You cannot search it, you cannot copy a line out of it, and a screen reader cannot voice it. OCR bridges that gap by recognizing the characters and attaching them as a real, selectable text layer.

That single step turns a stack of scans into documents you can search, quote, and reflow. It is also what makes a scanned PDF accessible to people who rely on assistive technology.

How to make a scan searchable

1. Open the OCR tool.
2. Add the scanned PDF or image.
3. Let it recognize the text on your device, then work with the searchable result.

Because the recognition happens locally, sensitive scans like IDs and medical records stay on your machine the whole time.

What makes OCR read a page well

Recognition works in stages. The software first finds the regions of the page that hold text, splits them into lines and words, and then matches each shape against a trained model of how letters look. The result is text with a position for every word, which is what lets a searchable PDF keep the picture of the page while hiding a text layer underneath it.

Most errors come from the source, not the software. A scan at around 300 dots per inch reads far better than a photo taken at an angle in dim light. Straight lines, dark text on a light background and a language setting that matches the page all help. Columns, tables and handwriting are harder, so check those parts of the result by eye before you copy figures out of them.