How to Extract Text from a PDF Page in Your Browser

OCR guide · 5 min read

You have a PDF open in your browser, you see the text right there on the screen, and you try to copy it — but nothing happens, or you get a garbled mess. Whether text pulls cleanly out of a PDF depends entirely on how the file was made, and once you understand the two kinds of PDFs, getting the words out becomes straightforward.

This guide covers how to tell which kind you have and how to extract text from either one, including the harder case: a scanned or image-only PDF where the "text" is really just a picture.

The two kinds of PDFs

Nearly every PDF falls into one of two categories, and they behave very differently when you try to grab text.

• Real-text PDFs. These were exported from a word processor, a web page, or design software. The characters are stored as actual text. You can select a word with your cursor, copy it, and search inside the document. Most invoices, reports, and ebooks are like this.
• Scanned or image PDFs. These were created by a scanner, a phone camera, or a screenshot. Each page is a flat image. The letters you see are pixels, not text — so selecting and copying does nothing useful. Old contracts, receipts, forms you photographed, and archived documents are usually like this.

How to tell which one you have

There is a one-second test: try to select a word. Click and drag your cursor across a line of text in the PDF.

• If the text highlights and you can copy it — you have a real-text PDF. Just select what you need and press Ctrl+C (or Cmd+C on Mac). You do not need OCR at all. Try this first, always.
• If nothing highlights, or your drag draws a selection box over the whole page like it is an image — you have a scanned or image PDF. This is where OCR comes in.

Extracting text from a scanned PDF page

When the text is baked into an image, the only way to recover it is optical character recognition (OCR) — software that looks at the picture and works out what letters are there. Textquill runs OCR directly in your browser, so you can point it at whatever PDF page is on screen and pull the text out.

The current approach is region-by-region and page-by-page. Here is how it works.

Step 1: Open the PDF in Chrome

Open the PDF file in Chrome's built-in viewer (just drag the file into a tab, or open it from a link). It works the same for a PDF embedded in a web page. Scroll to the page you want.

Step 2: Draw a box around the text

Press Alt+Shift+S to start an area selection, then drag a box around the text you want — a paragraph, a column, a single line, or the whole page. Textquill captures that region and extracts the text from it on your device.

Step 3: Repeat per region or per page

Move to the next block or scroll to the next page and select again. For a short document this takes a few passes; for a long one it is a manual, page-by-page process today. You can also capture the visible tab if you just want everything currently on screen.

Tips for cleaner results

OCR accuracy depends heavily on how clear the image is when you capture it. A few habits make a real difference:

• Zoom in first. Increase the PDF's zoom level before you select. A bigger, higher-resolution image gives the OCR more detail to read, which cuts down on misread characters. This is the single most effective tip.
• Do one column at a time. For newspapers, academic papers, or anything with multiple columns, select each column separately. Boxing across columns mixes the lines together and scrambles the reading order.
• Pick the right language. Make sure the OCR language matches the document. The wrong language setting causes accented letters and non-Latin scripts to come out wrong.
• Re-scan badly rotated pages. Straightening a skewed or sideways scan is not something you can fix in the browser. If a page is wildly tilted or upside down, a fresh, straight scan will read far better than fighting with a crooked one.
• Grab smaller regions. Selecting one clean paragraph at a time is often more accurate than trying to capture a busy full page in a single pass.

A note on privacy

PDFs are often the confidential ones — contracts, medical records, bank statements, signed forms. Many online "PDF to text" converters work by uploading your file to a server you do not control. Because Textquill performs OCR on-device, the page never leaves your computer. That matters when the document is sensitive, and it also means the extraction works offline.

An honest note on the current method

Right now this is a manual, region-by-region and page-by-page workflow: you open the PDF, select an area, and extract it. A dedicated feature that opens a PDF file and OCRs every page automatically is planned but not available yet. For a multi-page scanned document today, you repeat the area-select on each page. And remember the shortcut at the top — if the text selects normally, skip OCR entirely and just copy it.

FAQ