Search and Highlight Scanned English PDFs in Chrome (No Upload)

PDFs come in two flavours. Some carry their words as actual text — you can copy a sentence, search for a phrase, drag a highlighter through it. Others are pictures of pages: a scan, a photograph, a fax. There's no text underneath, only pixels. Open one in any reader and Ctrl+F gives you nothing.

Most online tools fix that by asking you to upload the document. For a draft NDA, an old medical record, or an internal report, that's a non-starter — privacy and confidentiality matter more than convenience. So Highlighty does it differently: when you open a scanned PDF, Highlighty reads the words right on your device. No upload, no third-party service, no waiting for a remote server. Once it's done, you can search and highlight the document like any other PDF.

TL;DR

• Open a scanned PDF in Highlighty's PDF viewer. If it's image-only, Highlighty reads the words automatically.
• Nothing leaves your device. The PDF stays in the tab, and so does the recognition.
• Each page becomes searchable as soon as it's done — you don't have to wait for the whole document.
• Mixed documents (some pages text, some pages scanned) work too. The text pages are searchable instantly; only the image-only pages need extra processing.
• Once a page is read, every Highlighty feature works on it: saved highlights, Cmd+F search, the matches sidebar.
• English in this first release. PDFs only — not images on regular web pages.

When this kicks in (and when it doesn't)

Highlighty doesn't run extra processing on every PDF. When you open a file, the viewer checks each page first. If a page already has selectable text, it's good to go and Highlighty doesn't touch it. Only pages that come up empty go through the extra reading step.

That gives you three behaviours:

• A normal text PDF: search works the moment you open it. No banner, no waiting.
• A fully scanned PDF: a small banner shows progress while pages are being read. Pages light up as they're done.
• A mixed PDF (common with old scans of new contracts, or hybrid forms): text pages are searchable right away; the scanned pages catch up in the background.

There's no setting to flip. The viewer makes the call per page.

Everything stays on your device

When you open a scanned PDF in Highlighty, the document never leaves the tab. The recognition runs inside the extension itself — no API call, no upload, no "anonymized" send-back, no remote server.

If you've ever stopped yourself from uploading a scanned contract, a medical document, or an internal report to a free online tool, that hesitation is exactly why this works the way it does.

What you'll see while it runs

When a scanned page is being processed, you'll see a small banner with a progress count: pages done, pages left, percentage complete. Each page becomes searchable as soon as it's done — you don't have to wait for the whole document. If your phrase is on page 3, you can usually find it before page 50 finishes.

Once a page is read, every Highlighty feature that works on a normal text PDF works on it: saved highlights light up across matches, the Ctrl+F bar finds and navigates them, and the sidebar lists matches by page.

Who this is for

If you regularly read documents that started life on paper, this was built for you. A few examples:

• Lawyers and paralegals working through scanned contracts, depositions, or discovery PDFs that can't be sent to third-party tools for confidentiality reasons.
• Researchers and historians reading scans of older books, journal articles, or archival documents where the original is the scan, not a typed file.
• Students working from textbook scans, lecture handouts, or older course readers that exist only as images.
• Journalists and policy analysts reviewing scanned reports, FOIA releases, or government documents where the source PDF is image-only.
• Anyone with a backlog of older internal reports, faxed-and-rescanned forms, or signed contracts that need to be searchable but can't leave the machine.

What if something goes wrong

Sometimes a scan is too damaged, or the file is set up oddly, and the reading step can't finish. When that happens, Highlighty puts up a small banner explaining that the document couldn't be read, with a retry button right there. Search still works on any pages that already had selectable text. If you switch to a different PDF mid-way, the previous one is dropped cleanly and the new one starts fresh.

Honest limits of this first release

Two things to know up front:

• English only in this first release. Other languages won't be read correctly. We'd rather ship one language well than several poorly.
• PDFs only. The reading step is wired into the bundled PDF viewer. Images on regular web pages aren't covered yet.

Quality also depends on the scan. Crisp 300-DPI scans give clean results. Documents that have been faxed, photocopied, and rescanned multiple times are harder for any reading tool, ours included.

Privacy summary

• Reading runs on your device. The PDF doesn't leave the tab.
• No remote API, no upload, no third-party service in the loop.
• No telemetry, no opt-in tracking, no usage reporting.

Frequently asked questions

Does Highlighty upload my PDF anywhere?

No. The reading happens inside the extension on your device. The PDF you open never leaves the tab, and no API call is made to a remote service.

What languages does this work with?

English in this first release. Other languages won't be read correctly. We chose to ship one language well rather than several poorly.

Does it work on images on regular web pages?

Not yet. This first release is wired into Highlighty's PDF viewer only. Images on standard web pages aren't covered.

How does Highlighty know which pages need reading?

When you open a PDF, the viewer checks each page for selectable text. Pages that already have text are searched directly. Pages that come up empty (the scanned ones) get the extra reading step.

How long does it take on a long scanned PDF?

It depends on your computer and the document length. Pages are processed one at a time in the background. The good news: each page becomes searchable as soon as it's done, so you can start searching the document well before it finishes.

What happens if a PDF has some pages with text and some scanned?

The text pages are searchable immediately and don't need any extra processing. Only the scanned pages go through the reading step, so mixed documents don't pay the full cost.

What if it can't read my document?

A banner appears explaining the document couldn't be read, with a retry button. Search still works on any pages that already had selectable text. If retry doesn't help, the file is likely too damaged or set up in a way no reading tool can handle.

Why didn't anything happen when I opened my scanned PDF?

If no progress banner appeared, the file probably has a real text layer underneath the scan — search works directly on it without the extra reading step. Try the Ctrl+F bar; you might find your text already.

Related reading

• How to search multiple keywords in a PDF in Chrome:

  Multi-term search across the bundled PDF viewer, including local files.

• A better Ctrl+F for Chrome:

  The same search bar works on PDFs you open through Highlighty.

• How job seekers use Highlighty:

  Pin keywords across job descriptions and PDFs of company materials.

• How to use Highlighty:

  Complete guide to search, highlight, X-Ray, and settings.

• Highlighty FAQ:

  Browser support, permissions, pricing, privacy, and webpage behaviour.

Try it

Update Highlighty (or install it for the first time from the Chrome Web Store or addons.mozilla.org), open a scanned PDF through the extension's viewer, and watch the progress banner. Search starts working on text pages immediately; the scanned ones light up as the reading catches up.

• Install Highlighty for Chrome:

  Chrome, Edge, Brave, Opera, and other Chromium browsers.

• Install Highlighty for Firefox:

  Mozilla Firefox add-on listing.