PDF to Image: Choosing the Right Resolution (DPI)

By PDF Hut team · Published September 27, 2026

Every PDF-to-image converter asks the same question: which resolution? Pick too low and text turns fuzzy; pick too high and you get huge files, slow conversions and, on a phone, a crashed browser tab. Here is how to choose with confidence.

Points, pixels and DPI

A PDF describes pages in points, a printing unit of 1/72 inch. An A4 page measures 595.28 × 841.89 points, a US Letter page 612 × 792. Most of what is on a page – text and vector drawings – has no resolution at all: it is a set of shapes that can be drawn at any size. Only embedded images, such as photos and scans, have a fixed number of pixels.

Converting a page to JPG or PNG means drawing it onto a grid of pixels. The DPI (dots per inch) setting decides how fine that grid is, measured against the paper size of the page:

pixels = page size in points ÷ 72 × DPI

So an A4 page (8.27 inches wide) rendered at 144 DPI is about 1190 pixels wide. DPI alone says nothing about quality – what matters is whether the pixel count is enough for where the image will end up.

The settings in PDF Hut

PDF to JPG offers three scale factors. Scale 1 maps one point to one pixel, which is 72 DPI; 2x and 3x double and triple that:

How many pixels do you actually need?

On screen

Screens don't care about DPI, only about pixels. A page shown full-width on a laptop or phone needs roughly 1000–2000 pixels across to look sharp, especially on high-density "Retina" displays. That makes 2x the right choice for most documents you share digitally. Use 1x for thumbnails and previews, where small files matter more than fine text.

In slides and documents

A presentation slide is typically 1920 × 1080 pixels when shown full screen. A page placed at half the slide width needs about 960 pixels across – covered by 2x. If the page image fills the whole slide height or you expect people to zoom into small print, use 3x.

For printing

Print is where DPI matters. For photos and fine detail, 300 DPI at the printed size is the common target; ordinary text documents look acceptable from about 150–200 DPI. The printed size changes the effective resolution: a 3x A4 image (216 DPI) printed at A5 – half the area – packs its pixels into a smaller space and reaches about 305 DPI. Printed at full A4 size, it stays at 216 DPI, which is fine for text but visibly soft for photos. For professional print work, use a desktop tool that renders at 300 DPI or more, such as Poppler's pdftoppm -r 300.

For text recognition (OCR)

OCR engines work best with enough pixels per letter. Tesseract's documentation recommends images of at least about 300 DPI. Within PDF Hut, 3x (216 DPI) is the closest option and is usually adequate for normal body text; small print and footnotes benefit from a higher-resolution desktop render. Choose PNG for OCR input, because JPEG artifacts around letters can confuse recognition.

JPEG or PNG?

JPEG compresses by discarding detail the eye is unlikely to notice. That works brilliantly for photographs and scans, which it usually stores in a fraction of the space PNG would need. Around sharp black-on-white edges, though, it leaves faint smudges. PNG is lossless: every pixel is kept exactly. For pages that are mostly text, charts and diagrams on a white background, PNG stays perfectly crisp and its files are often not much larger than JPEG. Rule of thumb: scans and photos → JPEG; digital text and graphics → PNG.

Resolution costs memory – squared

Doubling the scale doubles both width and height, so the pixel count grows fourfold, and tripling it grows ninefold. While a page is being drawn, the browser holds it as raw pixels at four bytes each: about 2 MB for an A4 page at 1x, 8 MB at 2x and 18 MB at 3x. PDF Hut renders every page of the document and keeps the finished images in memory for the previews, so a long document at 3x adds up quickly. On a phone, convert long documents at 2x, or cut out the pages you need first with Split PDF. Very large pages, such as posters and technical drawings, can exceed a browser's maximum canvas size at high scales.

The DPI number isn't stored in the image

An exported image is simply a grid of pixels. Images saved from a browser canvas usually don't carry a print-resolution tag, so some programs assume 72 or 96 DPI and report a surprisingly large physical size. That is only a label: when you place the image in a layout program or word processor, set the width you want and the pixels are spread accordingly.

Going back from images to PDF

JPG to PDF makes each page exactly the size of its image, at one point per pixel. A page rendered at 2x therefore comes back as a PDF page twice as wide and tall as the original. Viewers zoom to fit, and printing with "Fit to page" restores normal paper size – only the nominal page dimensions differ.

Summary

• Pixels = points ÷ 72 × DPI; what counts is having enough pixels for the destination.
• 2x (144 DPI) suits most screen use; 1x for thumbnails; 3x for printing smaller, zooming and OCR.
• JPEG for scans and photos, PNG for text and graphics.
• Memory grows with the square of the scale – be careful with long documents on phones.