PDF2OCR — Extract Text from Scanned PDFs & Images (Linux OCR)

Turn scanned PDFs, photos and screenshots into editable text using Tesseract OCR. Open files from your computer or straight from a URL, page through multi-page documents, translate the extracted text, and export it as plain text. Open source under the MIT licence.

Snap Store version GitHub stars
Snap Store GitHub

Download and install PDF2OCR

Linux — Snap

sudo snap install pdf2go

Run from source

git clone https://github.com/rishabh3354/PDF2OCR.git
cd PDF2OCR
python3 src/pdf2go.py

The application is MIT licensed — see the repository.

Features

Tesseract OCR

Accurate text recognition from scans, photos and screenshots.

Scanned PDFs

Extract text from PDF files with an unlimited number of pages.

Images too

JPG, JPEG, PNG, BMP, TIFF, GIF and WebP.

Open from a URL

Point it at a remote PDF or image instead of downloading first.

Page-by-page

Step through a multi-page document and see the text for each page.

Translate the output

Convert extracted text into another language in the app.

Export to .txt

Save the recognised text as a plain text file.

Caching

Previously processed pages load back quickly.

Runs offline

OCR happens locally — scanned documents never leave your machine.

Open source

MIT licensed, written in Python and PyQt5.

Supported input

Documents
PDF · scanned PDF · multi-page PDF
Images
JPG · JPEG · PNG · BMP · TIFF · GIF · WebP
Sources
Local files · folders · direct URL
Output
Editable text on screen · plain text (.txt) export · translated text

Common tasks

TaskWhat PDF2OCR does
Extract text from a scanned PDFOCR every page and read the text alongside the original.
Convert an image to textDrop in a JPG or PNG screenshot and pull the text out.
OCR a photographed documentWorks on phone photos of receipts, invoices and pages.
Get text from a PDF at a URLPaste the link — no separate download step.
Save the resultsExport the recognised text to a .txt file.
Translate a scanned pageExtract the text, then translate it in the app.
Digitise a book chapterPage through a long PDF and collect the text.

Screenshots

PDF2OCR extracting text from a PDF loaded by URL, original document on the left and recognised text on the right PDF2OCR extracting text from a local image file with the output language selector and page navigation

Frequently asked questions

Is PDF2OCR free?

Yes. It is free to install from the Snap Store, and the source is MIT licensed.

Does it send my documents anywhere?

OCR runs locally with Tesseract on your own machine, so scanned documents stay on your computer. The optional translate feature and opening a file by URL do use the network.

What file types can it read?

PDF files, including scanned and multi-page PDFs, plus JPG, JPEG, PNG, BMP, TIFF, GIF and WebP images.

Is there a page limit?

No. It handles PDFs with an unlimited number of pages.

Can it translate the extracted text?

Yes — the recognised text can be translated into another language inside the app.

How accurate is the OCR?

It uses Tesseract, so accuracy depends on the source. Clean 300 DPI scans read very well; low-resolution photos and unusual fonts read less well.

Can I export the text?

Yes, as a plain text (.txt) file.

Why is the snap called pdf2go?

The application shipped under the name PDF2Go before it was renamed, and the snap name cannot be changed without losing existing installs. `snap install pdf2go` is correct.

Is it available for Windows or macOS?

Not currently — the packaged build is Linux only. The source is Python and PyQt5, so it may run elsewhere, but that is untested.

Other apps by Warlord Softwares

AppWhat it does
FormatLabVideo & Audio Converter for Linux and Windows
4KTUBEYouTube Downloader for Linux and Windows
4KWALLWallpaper Manager & Auto Wallpaper Changer
DL-YouTubeMedia Downloader for 1400+ Sites
TUBE2GOOne-Click YouTube Downloader with System Monitor
JPEG2PDFImage to PDF Converter for Linux