How to Convert PDF Pages to Images (PNG or JPG) Without Uploading Your File
Reviewed by the FreeOnline.fyi team · Updated 2026-09-14
Why we built a converter that never uploads your PDF
We built this converter because the PDFs people actually need to turn into images tend to be the private ones: signed contracts, invoices, insurance letters, scanned IDs for an application form, lecture handouts with a student's name in the header. The usual workflow — pick a site, upload, wait in a queue, download — means a copy of that file now lives on somebody else's server, even if it is deleted later. PDF to Image Converter removes that step completely. Your file is read straight from disk into your browser's memory and rendered locally with PDF.js; there is no server endpoint involved, no login, and no queue.
What you notice in practice is what is missing. There is no upload progress bar, and no file-size ceiling imposed by a hosting plan. Speed now depends on your device instead of your connection, which is a real trade-off: a phone rendering a 60-page scan at 600 DPI will feel slower than a laptop doing the same job. Pages render one at a time onto a canvas, and each finished page is released before the next one begins, so memory stays roughly flat whether you convert a two-page form or a long report.
We kept the interface to a single screen on purpose: one drop zone, two option groups, one button. Nothing to configure before your first conversion, because the defaults (PNG at 150 DPI) already cover most document work.
Step by step: from PDF to a folder of page images
Step one is the drop zone. Drag your PDF onto the dashed box, or click the box and pick the file. The panel then shows the file name, its size, and the page count — worth glancing at the page count, because that number is the number of images you are about to create. The tool handles one PDF at a time; if you have three files, run them one after another rather than trying to stack them.
Step two is choosing output format and resolution. The defaults are PNG and 150 DPI, and if you are not sure, leave them. Click Convert and the button turns into a progress line reading "Processing page 3 of 12". That counter is honest, not decorative: pages are rendered serially, so the number only moves when a page is genuinely finished.
Step three is downloading. Finished pages appear as a thumbnail grid named page-1.png, page-2.png, and so on. Clicking a thumbnail downloads that page's full-resolution image — the thumbnail is only a preview, the actual file keeps the full pixel dimensions you selected. When the document has two or more pages, a Download ZIP button appears above the grid and packages everything as pdf-images.zip, assembled inside your browser with JSZip. Single-page documents skip the ZIP and just give you the one image.
PNG or JPG: decide by what is on the page, not by habit
PNG is lossless, which is why it is the default. Text, tables, thin rules, screenshots, signatures, and anything with hard edges stay crisp, and small text does not pick up the blocky artefacts that JPEG compression leaves around high-contrast edges. The W3C's PNG specification describes the format as a lossless, streamable image format — that losslessness is exactly what you want for a page you might zoom into later. The cost is file size: a text-heavy A4 page saved as PNG is usually several times larger than the same page as JPG.
JPG is the better choice when the page is photographic — a scanned photograph, a product catalogue, a brochure with gradients and backgrounds — or when you need to send twenty pages by email and care more about total size than pixel-perfect edges. We export JPG at a fixed quality of 0.92 and deliberately do not expose a quality slider; in our experience the slider mostly produces support questions, and 0.92 is a sensible balance between size and visible artefacts.
One behaviour worth knowing: pages are rendered onto a white canvas, so transparent areas in the source PDF come out white rather than transparent. For contract pages and scanned documents that makes no difference, but if you were hoping for a transparent-background PNG of a logo, this is not the right route.
Picking a DPI that matches the job
The four options map to a simple formula: the render scale is DPI divided by 72, because a PDF page is measured in points and 72 points equal one inch. That means an A4 page becomes roughly 595 × 842 pixels at 72 DPI, about 1240 × 1754 at 150 DPI, about 2480 × 3508 at 300 DPI, and about 4961 × 7016 at 600 DPI. US Letter is a little wider and shorter — around 2550 × 3300 at 300 DPI.
Use 72 DPI for quick previews, chat thumbnails, or a contact sheet of a long document. Use 150 DPI for slides, web pages, email attachments, and anything that will be viewed on a screen; it is the default because it is sharp enough for a full-screen zoom and small enough to stay manageable. Use 300 DPI when someone explicitly asks for print-quality images, or when a form says "attach 300 DPI scans". Reserve 600 DPI for the rare case where you need to crop a small region and keep detail — a single 600 DPI A4 canvas holds about 34.8 million pixels, roughly 139 MB of raw pixel data before compression, which is precisely why pages are rendered one at a time.
The most common misunderstanding is that a higher DPI adds detail. It cannot. If the source PDF contains a 150 DPI scan, exporting at 600 DPI produces a larger file with the same amount of real information in it, plus softer edges from interpolation. Match the output to the job rather than to the biggest number on the list.
Edge cases and mistakes we see most often
Password-protected PDFs will not open. The file has to be decrypted before PDF.js can read its pages, so either remove the password in a PDF reader first or export images directly from that reader. We would rather fail immediately than pretend to handle something we cannot.
Scanned documents behave differently from digital ones. Each page is already one big embedded image, so a 40-page scan produces 40 images that simply match the original scan resolution; converting at 600 DPI will not make a blurry scan readable. Big scans are also the heaviest jobs, and on a phone it helps to close other tabs before starting a long conversion. Pages with different sizes are another thing people miss: if page 1 is A4 and page 5 is Letter, page-1.png and page-5.png will have different pixel dimensions, which is expected rather than a bug.
Finally, keep the source PDF if you still need the text. Once a page is an image, the words inside it are pixels — not selectable, not searchable, and not extractable without OCR software. Think of the JPG or PNG as a picture of the page, not a replacement for it. And when the ZIP arrives, remember that file names inside are always page-1, page-2 and so on; if you need descriptive names, rename after extracting.
When this workflow fits, and when to reach for something else
This fits neatly when you need a page as a picture: dropping a contract page into a slide deck, sending a preview image to a client, uploading an image-only version to an application portal, or embedding a page in a product listing. The privacy angle matters most here — nothing leaves your device, so there is no upload to trust and no retention policy to read.
It is the wrong tool for OCR, PDF merging, splitting, or editing. Images are not searchable documents, and if someone asked for "an editable version", they need OCR, not a PNG. For those jobs, look through the other free online tools on FreeOnline.fyi rather than forcing this one to do something it does not claim to do.
One practical tip before you submit anything important: open the exported image and check its pixel dimensions in your image viewer. Some print shops and application portals judge "300 DPI" by pixel size rather than by DPI metadata, and PNG or JPG files do not always carry that metadata through. Our 300 DPI A4 output is about 2480 pixels wide — verify yours matches the requirement, and when in doubt, convert a single page first, check it, then run the whole document.