Background Remover

How Background Remover works

Make a plain background transparent with a colour flood fill from the edges. Tune the tolerance and download a PNG. Everything happens inside this page: the tool reads your input, runs the calculation or conversion locally, and hands the result back to you. Nothing is transmitted, because there is no backend to transmit it to.

That design choice is deliberate. Most free tool sites upload your files to a server, process them there, and quietly keep a copy. Toolcin cannot do that — the site is a set of static files. If your connection dropped the moment the page finished loading, every tool would still work.

How the removal works

The four corner pixels are averaged to estimate the background colour. Then a flood fill starts from every edge pixel and spreads inward, clearing any pixel whose distance from that background colour is within the tolerance you set.

Two details make this behave: the fill is four-directional from the edges, so an island of background colour inside the subject is not punched out; and the tolerance is a straight RGB distance, which is predictable and adjustable rather than a black box.

Choosing the tolerance

Too low and the background keeps a ragged halo where the JPEG compression shifted the white slightly. Too high and parts of the subject that happen to match the background disappear — a white shirt on a white background is the classic casualty.

Start at 28, raise it until the halo disappears, then stop. If the subject starts losing edges, the background is not uniform enough for this method.

What this cannot do, stated plainly

It is not an AI matting model. There is no understanding of hair, fur, glass or motion blur, and no edge feathering — the boundary is a hard cut at pixel level. On a person photographed in a room, the “background” is not one colour, so the fill stops almost immediately and nothing useful happens.

That is why the page says product shots and logos rather than portraits. A tool that promises a studio cut-out and returns a ragged outline wastes more time than one that tells you its limits first.

Getting a clean result

Shoot or choose an image with an even, plain background and reasonable contrast against the subject. PNG output keeps the transparency; JPEG would fill it with white. If the edge looks rough, a slight blur on the original before removal usually helps more than raising the tolerance.

Good to know

• Works offline once loaded, including on flights and in locked-down networks.
• No file size limit beyond your device's own memory.
• Free to use, with no account and no usage counter.
• Results are produced locally, so they are not logged or shared with anyone.

Questions