When to crop versus resize versus mask
Cropping cuts a rectangular section out of an image and discards everything outside it. The output is smaller in pixel count and shows less of the original scene, but the remaining pixels are exact — no resampling, no quality change. Use cropping when you want to remove content you don't need: trim background, fix composition, isolate a subject.
Resizing keeps the entire image but changes its pixel dimensions. The shape of the image (its aspect ratio) is preserved if you lock the ratio; the content is proportionally smaller (downscale) or proportionally larger (upscale, interpolated). Use resizing when you want the whole picture at a different size.
Masking is what you do when you want a non-rectangular cut — a circle, a rounded rectangle, an arbitrary shape. That requires either CSS at display time or a vector-aware editor at export time, not a pixel cropper.
Common aspect ratios and where they're used
- 1:1 (square) — Instagram feed default, Facebook profile photos, Twitter avatars, app icons.
- 16:9 (widescreen) — Most modern video (YouTube, TV, monitors), web hero banners, presentation slides.
- 9:16 (vertical) — TikTok, Reels, YouTube Shorts, Stories on every social platform, mobile video walls.
- 4:5 (Instagram portrait) — The tallest aspect Instagram allows in the feed without cropping; maximizes vertical space in mobile screens.
- 3:2 — DSLR and mirrorless camera sensor default. Standard photo prints (4×6, 6×9 inches).
- 2:3 (portrait) — Vertical photo prints, book covers, posters in portrait.
How to use this tool
- Drop a PNG, JPG, or WebP image onto the drop zone above.
- Pick an aspect ratio preset from the dropdown, or leave it on "Free" to crop arbitrary rectangles.
- Adjust X, Y, Width, Height to position the crop area. The red rectangle in the preview shows what will be extracted; everything outside it is darkened.
- Click Crop & download. The output is saved in the same format as the input.
Quality is preserved exactly
Unlike resizing, cropping doesn't resample the image — the output pixels are exactly the source pixels inside the selection rectangle, byte-for-byte. There's no quality loss from the crop itself. The only quality concern is the re-encode at the end: PNG sources stay lossless; JPG and WebP outputs are re-compressed at the widget's default quality, which is high enough to be visually identical to the source on almost any image.
If you absolutely need bit-perfect JPG output (rare — usually only forensic workflows care), use a dedicated JPEG-lossless-crop tool that operates directly on the JPG bitstream without re-encoding. That's a specialist use case; this tool covers everything else.