Turning Images into a PDF, Explained
How to convert JPG and other images to PDF: page size and orientation, margins, image order, why photo PDFs get large, and how to make a scanned PDF searchable.
By Udit Agarwal · Updated October 17, 2026
Converting images to PDF puts each picture on a page of its own and saves the pages as one document, so a set of JPGs, screenshots or phone photos becomes a single file that opens and prints the same way everywhere. Four choices shape the result: the page size, how each image is scaled onto its page, the order of the pages, and how the pixels are stored, which decides the file size. The pictures stay pictures, so the PDF has no text to search until OCR adds it.
PDFs measure pages in points of 1/72 inch. The common choices:
| Page size | Dimensions | In PDF points |
|---|---|---|
| A4 (ISO 216) | 210 × 297 mm | 595.28 × 841.89 |
| US Letter | 8.5 × 11 in | 612 × 792 |
| Fit to image | The picture's own shape | One point per pixel in ArrayKit |
A4 is standard in most of the world and Letter mainly in the US and Canada, so a PDF that will be printed should match the reader's paper.
Fit to image makes every page exactly the size of its picture, with no border, which looks best on screen. In ArrayKit each pixel becomes one point, so the nominal paper size is large: a 12-megapixel photo of 4032 × 3024 pixels makes a page of 56 × 42 inches (about 142 × 107 cm). Viewers zoom it to fit the window, and printing it needs the print dialog's fit-to-page option.
ArrayKit's Image to PDF offers fit to image (Auto) and A4, but not Letter. An A4 page prints on Letter paper at about 94 percent with fit to page, and Scan to PDF can make Letter or Legal pages directly.
A photo on a page of the other orientation, such as a wide photo on a tall page, leaves a lot of empty paper, because its long side has to fit the page's short side. Here is a 4032 × 3024 photo placed on A4 by Image to PDF:
| Page | Printed size of the photo | Margins |
|---|---|---|
| Portrait A4 | 193 × 145 mm | 8.4 mm at the sides, 76 mm above and below |
| Landscape A4 | 258 × 193 mm | 19.7 mm at the sides, 8.4 mm above and below |
On the wide page the photo prints a third larger in each direction. Portrait and Landscape give every page the same orientation, so a mix of tall and wide photos suits Auto, which follows each picture's shape.
On a fixed page, the sensible approach is to scale the image until it fits inside the page, keeping its proportions so nothing is cropped or stretched. Image to PDF does this in its A4 modes, then shrinks the image to 92 percent of that size, which leaves a margin of 4 percent of the page on the two edges where the fit is tightest: 8.4 mm across the 210 mm side of A4, or 11.9 mm along the 297 mm side. There is no margin setting, and Auto pages have no margin at all. For a binding edge or room for notes, Add Margins to PDF adds white space to finished pages.
Scaling to fit also works upwards. A small image is enlarged to fill the page, and enlarging adds no detail: a 400-pixel-wide image spread across the 193 mm of a portrait A4 page prints at about 53 pixels per inch and looks soft. Sharp printing needs roughly 300 pixels per inch at the printed size, which for a photo across a portrait A4 page means about 2,300 pixels wide.
Pages follow the order of the images. Image to PDF adds them in the order you choose them, numbers each preview, and lets you move any image left or right, remove it or add more at the end. Files selected together arrive in the order your browser hands over, often alphabetical but not always, so name them to sort the way you want (01-cover.jpg, 02-receipt.jpg) and check the numbers before you create the PDF.
The PDF is named after the first image, so a set that starts with 01-cover.jpg downloads as 01-cover.pdf.
A converter that runs in the browser can only use images the browser itself can open:
| Format | Works? | Notes |
|---|---|---|
| JPG, PNG, WebP, BMP | Yes | Any transparent areas show as white paper |
| GIF | Yes | Only the first frame of an animation |
| AVIF | In current browsers | |
| HEIC (iPhone photos) | Usually not | Convert with HEIC to JPG first |
| TIFF | Usually not | TIFF to PDF converts it directly, multi-page scans included |
| SVG | Usually not | SVG to PDF keeps shapes as vectors where it can |
Files that are not images at all, such as a document dropped by mistake, are skipped with a notice. An image your browser cannot decode, such as a HEIC photo in Chrome, is left out of the PDF and named in a notice, so you can convert it and add it again.
A converter can put a JPG into a PDF in two ways: copy the JPG data in unchanged, which keeps the file small, or decode the picture and store its pixels again.
Image to PDF does the second, at full pixel size, and the format decides how the pixels are stored. JPG photos are saved again as JPEG at 92 percent quality, so a photo PDF stays close to the photo's own size: a 1920 × 1080 photo saved as a 431 KB JPG made a one-page PDF of 433 KB, and a 5.6 MB, 3840 × 2160 JPG saved at a higher quality came out at 1.35 MB. PNG, WebP, GIF and BMP images are stored losslessly as PNG, which keeps screenshots, scans and graphics exact but makes a photo in one of those formats large: the same 1920 × 1080 picture as a 2.2 MB PNG made a 2.9 MB PDF.
Choosing A4 does not change this. The photo is drawn smaller but keeps all its pixels, so a 4032 × 3024 photo across a portrait A4 page sits at about 530 pixels per inch, far more than print needs, and the file is as large as with Auto.
Fewer pixels is the dependable fix. Halving the width and height leaves a quarter of the pixels: the 3840 × 2160 photo above made a 1.35 MB PDF, and the same photo resized to 1920 × 1080 made one of 433 KB. For photos, JPG input also beats PNG, because PNG pages are stored losslessly. For reading on screen, about 2,000 pixels on the long side is plenty, and for print, use the 300 pixels per inch rule above. Image Resizer shrinks a whole batch of photos at once.
If the PDF is still too big, Compress PDF redraws each page as a JPEG. On pages made with Auto, its default Medium setting keeps every pixel, because the page has one point per pixel, and it saves the page again at 60 percent JPEG quality, which takes a lossless PNG page down to a fraction of its size. On A4 pages the same setting draws the whole page at 595 × 841 pixels, which leaves a photo only about 550 pixels across, so build with Auto if you plan to compress. Why PDFs get big covers compression and email limits in depth.
A phone photo of a document makes a poor scan until it is cleaned up. Crop it to the edges of the paper with Image Cropper, and take it straight from above, in even light, without your shadow on the page. For flatbed scans of documents, 200 to 300 DPI is plenty.
Scan to PDF is built for this job. It takes pages from your camera or from photos you drop in, can turn them greyscale or high contrast so text stands out, stores each page as a JPEG, which keeps the PDF small, and offers A4, Letter, Legal or fit-to-photo pages. If scanned pages come out tilted, Deskew PDF measures each page's angle and straightens it.
A PDF made from images holds pictures of text, not text, so you cannot search it, copy from it or have it read aloud. Optical character recognition (OCR) reads the letters in each picture and adds an invisible text layer under it. The page looks the same, but search and copy now work.
PDF OCR does this to a finished PDF. Choose the language on the page from the 12 offered, among them English, Spanish, Hindi, Chinese and Arabic, and it returns a searchable PDF and a plain-text copy, up to 30 pages per run. Recognition is only as good as the picture, so straight, sharp, high-contrast pages work best, and names and numbers are worth checking by eye. For the text of a single image without a PDF, use Image to Text.
Image to PDF combines images into one PDF. Drop or choose images at any time, put them in order with the numbered previews, pick Auto (fit image), Portrait or Landscape, then Create PDF. Auto sizes each page to its image, and Portrait and Landscape centre each image on an A4 page at 92 percent of the largest size that fits. The download is named after the first image.
Its limits follow from how it works: images keep their full pixel size, and PNG, WebP, GIF and BMP pages are stored losslessly, so photos in those formats make large PDFs; there is no Letter size, margin setting or compression option; and images your browser cannot open, typically TIFF and HEIC, are left out with a notice and need converting first. Everything happens in your browser on your own device, and the images are not uploaded.
How do I combine several photos into one PDF? Add all the images to a converter such as Image to PDF, put them in order, choose a page size and create the PDF. Each image becomes one page, in the order shown.
Does converting JPG to PDF reduce quality? Not with Image to PDF, which stores each image at its full pixel size without further lossy compression. Other converters may recompress or downscale, and compressing the finished PDF does give up some detail.
Why is my PDF bigger than the photos? JPG photos are saved again as JPEG, so they stay close to their own size, but PNG, WebP, GIF and BMP images are stored losslessly, and a photo in one of those formats takes a lot of space. Use JPG for photos, resize them to about 2,000 pixels on the long side before converting, or compress the finished PDF.
Can I make a Letter-size PDF from images? Image to PDF offers fit-to-image and A4 pages only. Scan to PDF offers Letter and Legal, and an A4 PDF prints on Letter paper at about 94 percent with fit to page.
Can I convert HEIC or TIFF images to PDF? Most browsers cannot open them, so convert them first. HEIC to JPG handles iPhone photos, and TIFF to PDF turns TIFF scans, multi-page ones included, into a PDF directly.
Will the PDF be searchable? No. A PDF made from images holds pictures of text. Run it through PDF OCR to add a searchable text layer.
Are my images uploaded? Not with ArrayKit's Image to PDF. The images are decoded and the PDF is built in your browser, on your own device.