Z

Image Watermark

Add a text watermark to images locally and download the result. Nothing is uploaded.

Runs in your browser — files never leave your device

How it works

Drop an image, type your text, and set an opacity. The tool draws the photo onto a canvas at its full original resolution, then renders the text twice in the bottom-right corner — first a dark outline, then a white fill on top — so the mark stays readable on light and dark backgrounds alike. The preview updates live as you type or move the slider, and the download reproduces it exactly.

Text size is proportional to the image: bold sans-serif at the image width divided by 18, never below 16px. On a typical 4032 × 3024 phone photo that works out to 224px letters inset 112px from the right and bottom edges — about the visual weight of a standard photo credit. The outline stays thin (the font size divided by 16, so 14px on that photo), reading as an edge rather than a border.

The opacity slider (10–100% in 5% steps, default 40%) controls the white fill directly, while the outline is drawn at 60% of the same value so the two fade together. Around 30–50% gives the classic translucent watermark that identifies an image without hiding it; push toward 100% for proof images where you want reuse to be genuinely unattractive.

The image never leaves your machine — compositing happens on an in-browser canvas and the file downloads from memory, so unpublished or client-confidential images are safe here. Keep the text short: it’s a single line anchored to the corner, so a long sentence on a narrow image can run past the left edge. The output is a PNG at full resolution with no EXIF metadata, named after the original with a watermarked- prefix.

Frequently asked questions

Can I move the watermark or tile it across the image?
No — placement is fixed in the bottom-right corner, inset by half the text height, and the text is drawn once on a single line. There’s no tiling or drag-to-position; the tool is built for the common case of a small corner credit.
How big will the text be?
The font size scales with the image: the width divided by 18, with a 16px minimum so it stays legible on tiny images. On a 4032-pixel-wide photo that’s 224px text inset 112px from the corner; on a 1920-pixel-wide image it’s 107px. Because it’s proportional, the mark occupies a similar share of any image you feed it.
Why is the text white with a dark outline?
A plain white mark disappears on light skies and a plain black one disappears in shadows. Drawing a white fill over a black outline — the outline at 60% of your chosen opacity — keeps the text readable on almost any background without you having to pick colors per photo.
What opacity should I choose?
The default is 40%, which reads clearly without dominating the photo. Go lower (10–25%) for a subtle credit on busy images, and higher (60–100%) when deterrence matters more than looks — proofs and previews you don’t want reused as-is.
Does a watermark actually stop image theft?
It deters casual copying and keeps attribution attached as the image spreads, but it isn’t protection: a corner mark can be cropped off or cloned out with editing tools. For images you sell, watermark the previews prominently and deliver clean files only after purchase.
What format is the output, and is anything else changed?
The result is always a PNG at the original pixel dimensions, with transparency preserved if the source had any. Like all canvas output it’s a brand-new file, so EXIF metadata — camera details, capture date, GPS — is not carried over. A photo re-encoded as PNG will usually be a larger file than its JPG original.
Is my image uploaded to add the watermark?
No. The photo is drawn onto a canvas in your browser, the text is rendered on top, and the download is produced on your device — nothing is transmitted or stored anywhere. You can watermark client work, unreleased product shots, or personal photos without them ever touching a server.

Related tools