Crop PDF

What is the CropBox?

A PDF page is described by several rectangles, defined in the PDF 1.7 specification. The two that matter for visible-area cropping are:

• MediaBox — the physical page boundary. Every page must declare one. This is what a printer thinks of as “the paper”.
• CropBox — the region of the MediaBox that viewers and printers should render. If a page has no CropBox, viewers fall back to the MediaBox.

The other three (BleedBox, TrimBox, ArtBox) exist for print production and don’t change what a normal reader sees. Cropping for on-screen reading and ordinary printing is, almost always, a CropBox edit.

Because the CropBox just narrows the visible window, the underlying content stream — text, images, vectors, annotations — is unchanged. The cropped pixels still exist in the file; they’re just hidden. This is what makes PDF cropping non-destructive and reversible: any tool that can reset the CropBox to match the MediaBox restores the original view.

What this tool does

You drop a PDF, enter four margin values (top, right, bottom, left), optionally pick a page range, and the tool writes a smaller CropBox on each selected page. Margins are in PDF points — 72 points equals 1 inch, which means 25.4 mm. A4 paper is 595 × 842 points; US Letter is 612 × 792 points.

Internally, for each page in the range, the tool reads the current MediaBox, computes the new CropBox as (left, bottom, width − right, height − top), and calls pdf-lib’s setCropBox(x, y, w, h). The PDF is then re-serialised and downloaded.

Every step runs locally in the browser tab.

Why margins, not coordinates

The PDF coordinate system has its origin at the bottom-left of the page and Y increases upward — a convention inherited from PostScript. Asking a user to enter “x=36, y=42, width=523, height=758” forces them to do mental arithmetic against the page dimensions for every page size they touch.

This tool instead asks for per-side margins: the strip you want to trim off the top, right, bottom, left. That maps directly to what a human sees on the page (“there’s a scanner-shadow strip about 18 points wide on the left”) and works identically on any page size, since the tool reads the current MediaBox and computes the inset rectangle itself.

Common use cases

• Removing scanner artefacts. Documents scanned on flatbed or sheet-fed scanners often have a few millimetres of black, grey, or shadow at the page edges. A 12–18 point margin on each side hides them cleanly.
• Tablet-friendly reading. Academic PDFs with generous print margins waste a quarter of a tablet’s screen. Cropping 30–50 points off the left and right and 50–70 off the top and bottom enlarges the body text dramatically without re-flowing the layout — useful for textbooks, papers, and standards documents on iPad or Kindle Scribe.
• Trimming printer marks and bleed. Press-ready PDFs include crop marks, registration marks, and a bleed area outside the finished page. For sharing the file as the finished document rather than as production artwork, crop those off.
• Reformatting scanned books. Public-domain book scans on Archive.org and Google Books frequently have the gutter (binding shadow), page edges, and library stamps in the margin. A few passes of cropping cleans the pages without re-OCRing the text.
• Hiding letterheads or footers. A consistent letterhead band or page-number footer that isn’t wanted for archival cropping is one CropBox edit away.

How to use this Crop PDF tool

1. Drop a PDF onto the dropzone. The first page renders as a preview with its dimensions in points.
2. Enter the four margins — top, right, bottom, left — in PDF points. Start small (10–20) and re-run to fine-tune.
3. Optionally, enter a page range (e.g. 1-5 or 1-3, 7, 10-12). Leave blank to crop every page.
4. Click Crop PDF. The tool writes the new CropBox on each selected page and produces a downloadable Blob.
5. Open the downloaded file in your reader of choice. If the crop is too aggressive, re-run with smaller margins on the original PDF — cropping is non-destructive, but each pass narrows the visible window further if you re-crop the cropped output.

Security and limits considerations

• The original content stream is preserved. Anyone with a PDF editor could open the file and reset the CropBox to the MediaBox, exposing the hidden margins again. If your goal is to remove sensitive information on the page edges (not just hide it visually), cropping is the wrong tool — use Redact PDF, which rasterises the page so the underlying pixels are destroyed.
• Page rotation (a separate page-dictionary entry) is respected: the margins you enter refer to the visual top / right / bottom / left of the page as the reader sees it.
• Encrypted PDFs need to be unlocked first. Use PDF Password Remover before cropping if the file is password-protected.

Privacy

The PDF is opened, modified, and re-saved entirely in this browser tab. There is no upload, no temporary storage, no telemetry on file contents. The Cloudflare edge serves the static page assets; after they load, only your CPU is at work.

Compatibility notes

The output is a standard PDF whose only change is a smaller CropBox on the affected pages. Every modern reader honours the CropBox: Adobe Acrobat, Apple Preview, the in-browser viewers in Chrome, Edge, Firefox and Safari, the Books app on iOS, Drawboard and Xodo on Windows. Printing clips to the CropBox on essentially all consumer printers. The exception is some legacy print- production workflows that intentionally render the full MediaBox to expose crop marks — irrelevant for everyday cropping.