Image Cropper

Cropping is the one image edit almost everyone needs and almost every tool gets slightly wrong: the dragged frame lands a few pixels off, the exported size is not what the numbers said, or the whole thing happens on a server that has now seen your photo. This cropper keeps a single source of truth — the crop rectangle in image pixels — and lets you set it either by dragging or by typing. Both paths write the same four numbers, so the result is the same either way.

The work happens in your browser. The region you keep is copied out of the decoded image onto a fresh canvas and encoded into the format you choose, and the download is the only thing that leaves the tab.

1. 1Load the image. Drop a file or browse for one. The full picture is shown with the crop frame covering all of it, so the starting point is “keep everything”.
2. 2Choose a shape, or stay free. Free lets every edge move independently. The ratio presets hold a shape while you resize the frame from any corner.
3. 3Position the frame. Drag the body to move it, drag a corner handle to resize it, or use the arrow keys to nudge by one pixel — Shift for ten. The numeric fields update as you go.
4. 4Or type the numbers. X, Y, width and height accept exact values and clamp themselves to the image, so you cannot ask for a region that runs off the edge.
5. 5Download the region. The output line shows the exact dimensions and the encoded size. The file is named with the new size so it is easy to find later.

Ratio

Free, 1:1, 4:3, 3:2, 16:9 or 9:16. Picking a preset keeps the frame’s current width, derives the height from the ratio, and keeps the frame centred where it already was. “Centre the frame” then fits the largest frame of that shape on the image.

X and Y

The top-left corner of the crop, in pixels from the image’s top-left corner. Values beyond the image edge are pulled back inside.

Width and height

The size of the kept region. The minimum is 16 pixels on each side; the maximum is whatever fits from the chosen corner.

Output format

Keep the source format or write JPG, PNG or WebP. Only formats the browser has proven it can encode are listed.

Quality

Appears for lossy formats. It starts at 88 here because a crop is usually a finishing step and you rarely want to lose much detail at this point.

Any image the browser can decode can be cropped, including JPG, PNG, WebP, GIF, BMP and in most browsers AVIF. Output is JPG, PNG, WebP, or AVIF where the encoder test passes.

• The crop is rectangular. Circles, rounded corners and freehand shapes are not something this tool draws — crop to a square and apply the mask in the tool that needs it.
• Rotation is a separate step. If a horizon needs levelling, use the rotate tool first; this cropper does not rotate by arbitrary angles.
• A cropped JPG is re-encoded. Writing PNG avoids that loss at the cost of a larger file.
• Metadata is not carried through. A cropped photo has no EXIF, GPS or colour profile tags.
• Images beyond roughly 100 megapixels are refused up front because the canvas cannot usually be allocated.
• Animated files are cropped on their first frame only.

A 4032 × 3024 photograph is 4:3. Choosing 16:9 drops a 4032 × 2268 frame across the full width, centred vertically. Dragging it up or down picks which band of the picture survives; the width stays at 4032 because there is nowhere wider to go. Downloading at this point gives a 16:9 image at full resolution.

If the destination needs exactly 1280 × 720, that is a resize rather than a crop: send the result through the image resizer with the width set to 1280 and the ratio locked. The height will come out at 720 because the shape is already correct.

• Drag or type — same result

  The frame and the number fields describe one rectangle. Whatever you set last wins, and the other view follows exactly.

• Pixel-exact

  The output is the region you asked for, to the pixel. The readout reports the actual dimensions of the encoded file.

• Keyboard operable

  The frame can be focused and nudged with arrow keys, and every field has a label, so the tool works without a pointer.

• No upload delay

  The picture is decoded once on your device; moving the frame is instant because nothing has to be recomputed on a server.

• No account, no watermark

  Nothing gates the download. The tool is the whole product.

• Private by construction

  The tool has no upload path at all, so there is no server-side copy to worry about deleting later.

Nowhere. It is decoded into this tab’s memory, displayed, and the selected region is copied to a new canvas and encoded there. No request carrying the image is made, and no part of it is written to local storage, a cache or a cookie. Loading another file or closing the tab releases it.

The page itself sets no cookies and runs no analytics. The server that served it can log the page request — as any web host can — and the Privacy Policy describes exactly what that means.

PixelEditor is a small set of single-purpose image tools that run inside your web browser. Each one does a specific job — resize, crop, rotate, compress, convert, inspect, strip metadata — and each one has its own page so you can bookmark exactly the tool you keep needing.

The tools are built with the Canvas API and Web Workers. When you pick a file, the browser reads it from your own disk, works on it in memory, and hands you the result back as a download. There is no processing server involved, so there is no upload step, no queue, and no account.

Everything is free, there are no watermarks, and there is no paid tier that unlocks the real version. Read more about how PixelEditor works and why.