Image PDF Converter

About Image PDF Converter

This tool bundles several images into one PDF, or pulls the JPEG images out of a PDF. The direction follows what you pick: images produce a PDF; a PDF produces images.

The important detail is that neither direction redraws anything. Images to PDF places a JPEG photo's own bytes directly inside the PDF, and encodes other formats as JPEG first. PDF to images finds image streams that are stored as JPEG inside the PDF and saves them byte for byte. That makes it fast and avoids an extra round of quality loss, but it also means a PDF page made of text and vector shapes is not turned into a picture.

Page size comes from the image's DPI, or else from the paper

PDF measures in points, 1/72 of an inch. An image file only has pixels, so its physical size depends on how many pixels make an inch (its DPI). This tool reads the DPI recorded in the file (JFIF or EXIF in a JPEG, pHYs in a PNG) and computes width in points = pixels ÷ DPI × 72. A Letter page scanned at 300 dpi (2550 × 3300 pixels) comes out as 612 × 792 pt, exactly 8.5 × 11 in, so printing at Actual size doesn't tile or crop.

Phone photos often carry no DPI, or only the 72 or 96 dpi placeholder that cameras and editors fill in; taken literally, that would make pages over a meter tall. So anything below 100 dpi counts as missing, and the image is fitted to paper instead: Letter when your browser language is US English (en-US), A4 otherwise. A wide image gets a landscape sheet; the image is scaled up as far as its proportions allow and centered, with white at the edges.

If a portal insists on letter-size or A4 pages, choose Fit to Letter or Fit to A4 under Page size. Images that have a DPI are fitted to that paper too.

What changes going from images to PDF

• JPEG photos are not re-encoded: their compressed data goes in as it is, so there's no quality loss and the size stays close to the original. Only JPEGs a PDF can't hold directly (CMYK JPEGs, for example) and other formats such as PNG or WebP are drawn onto a canvas and encoded as JPEG at quality 0.9.
• Transparent areas of a PNG are filled with white, because JPEG has no transparency, the same as printing it on white paper.
• Color is stored as RGB (DeviceRGB), and grayscale JPEGs as DeviceGray. A printer that wants CMYK needs a separate conversion.
• Phone orientation (the EXIF rotation tag) is applied on the page so photos stand upright. Even for JPEGs that go in unchanged, the EXIF and XMP segments (camera details, GPS location) are removed first.
• Pages follow the list order; drag to reorder.

Which PDFs give images back, and which don't

The PDF-to-images direction looks for image objects using the DCTDecode (JPEG) filter and saves their bytes unchanged. There is no re-encoding, so no quality loss, and it works best on scanner output and on one-photo-per-page PDFs like the ones this tool creates.

A document exported from a word processor is mostly text and vector drawing, so there's nothing to extract and those pages are skipped. Flate-compressed (PNG-style) images, JPEG 2000, and the CCITT and JBIG2 formats common in black-and-white scans are not extracted either, and image dictionaries packed inside compressed object streams may be missed. If you need a picture of each page as it looks, use a tool that renders pages, such as the PDF-to-long-image converter. CMYK JPEGs are extracted as they are, so some photo viewers show their colors inverted.

Worked examples

How to use it

1. Choose several images or one PDF.
2. For images, drag to set the order.
3. Convert and download a single PDF or a ZIP of images.

Key terms

Point (pt)

PDF's length unit, 1/72 inch. Letter is 612 × 792 pt.

DCTDecode

The PDF filter name for JPEG-compressed images; the stream bytes are a JPEG file.

ppi / dpi

How many pixels map to an inch. Changing it doesn't change the pixels, only the nominal physical size.

Frequently asked questions