Photos of paper and scanner output come out grey, dim, and framed by whatever the page was lying on. This pushes the background back to white and crops away the desk.
Put the before and after side by side before you keep it.
This one needs an upload, and is deleted afterwards
Browsers cannot do this calculation, so the file goes to our server. It is deleted automatically 60 minutes after the job finishes, and the working copy lives in a memory filesystem that is wiped whenever the server restarts.
We do not log file names and do not log file contents — that is a hard rule in our engineering standards, not a setting.
What it does — and what it doesn't
It fixes lighting and background colour. It does not touch geometry. A crooked page goes in crooked and comes out crooked — straightening means repainting every pixel, and on a document you may need as evidence that is retouching. We deliberately do not.
Pages with no scanned image are passed through untouched, so the text layer is never destroyed.
Pages that are one big dark photograph are skipped rather than bleached.
Creases, folds, and shadows falling across the letters cannot be fixed here — those need a better photograph.
Common questions
How is this different from just raising the contrast?
Raising contrast pulls the whole image at once, so dark corners get darker and letters fill in. This estimates how the paper tone varies across the page first, so unevenly lit photos do not end up half black.
How much of the desk gets cropped?
It finds the sheet of paper and takes the largest rectangle inside it. If an edge of the paper runs off the frame, it works with what is visible.
Can I run OCR on the result?
Yes, and you should — recognition on clean white is far better than on grey.
