Watermark Image

The image is drawn onto a canvas and the text painted over it, then the whole thing is re-encoded. The result is a single flat picture: the watermark is part of the pixels, not a layer sitting on top, so it survives being copied, resized and re-saved.

The text is sized relative to the image, not in fixed pixels. A 5% setting means five percent of the shorter side, so the same setting looks the same on a 600px thumbnail and on a 6000px photograph. Fixed point sizes are the usual reason a watermark looks right on one image and microscopic on the next.

A soft dark shadow is drawn behind the text automatically. Without it, white lettering disappears against a bright sky and black lettering vanishes into a shadow — the shadow keeps the mark legible over any background without needing a solid box behind it.

Everything happens on your device, and the preview re-renders as you type, so what you see before downloading is the actual output rather than an approximation.

You are publishing sunset.jpg, 3000×2000, to a portfolio and want a discreet credit line in the corner.

Choose the file and type © 2026 Baker Alazzam. Leave the position on Bottom right. At the default 5% the text renders at 100px — five percent of the 2000px short side — which on this image is noticeable but not shouting.

The sunset is bright, so white text is fighting the sky. Two options: drop the strength to around 55% so it reads as a subtle mark, or switch the colour to black, which the automatic shadow keeps legible. Try both in the preview; on a warm-toned photograph white at 55% usually wins.

Now the case where the defaults fail. Drag the size to 2% and the credit becomes a barely visible smudge on a 3000px image — technically present, useless as a credit. Drag to 14% and it dominates the frame. The workable range for a corner credit is 3–6%; 10% and above is for a deliberately obtrusive mark across the middle.

One thing this does not do: protect the image. A watermark in the corner can be cropped off in seconds. If deterrence is the goal, centre it and raise the strength, and accept that you are trading the look of the photograph for it. If the photo is going online, also strip its EXIF data — a credit line in the corner is no use if the file quietly carries your home coordinates.