Crop an image

Cut away the empty margin, the accidental thumb in the corner, or everything except the one chart you needed from a screenshot. Give the offset and size of the region to keep and slimdocs extracts exactly those pixels — no re-encoding of anything outside the crop.

How to use it

1. Open the Crop Image tool and drop in your photo.
2. Enter the left and top offset of the region you want to keep, in pixels from the top-left corner.
3. Enter the width and height of that region — leave them blank to run to the edge.
4. Crop, then download the result.

How the crop box is measured

Every crop is described by four numbers: how far in from the left edge to start, how far down from the top to start, and then how wide and how tall the kept region should be. All four are in pixels, and the origin — 0, 0 — is the top-left corner of the image, which is the convention every image editor uses.

Leaving the width or height blank means "run to the edge from where I started". So a left of 200 with a blank width keeps everything from 200 pixels in to the right-hand edge. If the box you describe falls outside the image, slimdocs tells you the actual dimensions rather than silently clamping the numbers and handing back something you did not ask for.

Cropping is not the same as resizing

It is worth being clear about the difference, because the two are easy to confuse. Resizing keeps the whole picture and changes how many pixels it is drawn with — everything stays in frame, just smaller or larger. Cropping keeps the pixels at their original scale and throws away everything outside the box.

That means cropping never softens what remains. A cropped region is bit-for-bit the same detail it was in the original, which is why cropping first and resizing second gives a better result than doing it the other way round.

When cropping is the right tool

• Trimming a screenshot down to the single dialog, chart or error message that matters.
• Removing a scanner's black border or the desk visible around a photographed document.
• Turning a landscape photo into the square an avatar field demands, choosing which part survives.
• Cutting a person or object out of the edge of a group shot before sharing it.
• Removing a burned-in caption, watermark strip or timestamp along one edge.

Orientation is handled first

Phone photos usually record their rotation as metadata rather than rotating the actual pixels. That creates a real trap for cropping: the coordinates you measured while looking at the upright image would land somewhere else entirely on the unrotated pixel data.

slimdocs applies the orientation before it reads the dimensions or extracts the region, so the numbers you type always refer to the image as you see it. The dimensions reported in an out-of-bounds error are the upright ones too.