How to Crop an Image Online for Free (No Upload Required)

Cropping is the most common image edit. You need a passport photo at exactly 35×45mm, a LinkedIn headshot at square ratio, a product image at 4:3, or just to remove empty space from the edges of a screenshot. Most free online tools handle this fine — but they upload your photo to a server to do it. For casual screenshots that is not a concern. For photos that contain location data, recognisable faces, or anything sensitive, it is worth knowing you have a faster, more private alternative.

How to crop an image in your browser — step by step

1. Open keptlocal's Crop Image tool.
2. Drop your image onto the upload zone, or click to browse. JPG, PNG, and WebP are all supported.
3. A selection box appears on the image. Drag the corners and edges to define the area you want to keep. Move the selection without resizing by dragging from its centre.
4. To constrain to a fixed ratio — square, 4:3, 16:9, or passport — select it from the ratio menu. The selection will snap to that ratio as you resize it.
5. Click Crop. The cropped image downloads immediately. Nothing is sent to a server.

You can verify this yourself: open DevTools (F12) → Network tab, then crop an image. You will see the initial page load and nothing further — no upload, no external request during processing.

Free-form vs. fixed-ratio cropping

Free-form cropping lets you draw any rectangular selection without constraint. Use this when you are cropping to remove content rather than to meet a dimensional specification — trimming a border from a scan, removing a distracting element from one side of a photo, or cutting a specific region out of a screenshot.

Fixed-ratio cropping locks the selection to a specific width-to-height ratio. The selection stays at that ratio as you resize it — dragging a corner makes the box larger or smaller while keeping the proportions constant. Use this when the image needs to fit a specific format: a social media post, a form photo, or a template slot with defined dimensions.

Common crop ratios and when to use them

• 1:1 (square) — Instagram posts, profile photos, avatar icons, product thumbnails on square-grid layouts. Virtually every social platform uses square images somewhere in its interface.
• 4:3 — traditional photo format, most monitor displays before widescreen became standard. Still widely used for product photography and web layouts where images sit inside cards.
• 16:9 — widescreen video thumbnails (YouTube, Vimeo), presentation slides, blog post featured images on most modern themes. If you are making a cover image for a YouTube video, this is the ratio to use.
• 3:2 — the native ratio of most DSLR cameras and the 35mm film format. If you are cropping a DSLR photo and want to maintain the original proportions while just removing the edges, 3:2 keeps the image looking like it was shot at that size.
• A4 (approximately 1:1.414) — useful when preparing an image to fill a page that will be printed at A4. Ensures the image fills the page without distortion or cropping at print time.
• Passport (35×45mm, approximately 7:9) — the standard size for passport and visa photos in most countries. The face should occupy 70–80% of the frame height. Note that official passport photos have additional requirements (neutral background, specific lighting, no shadows) that cropping alone does not guarantee — check your country's specification before submitting.

Getting precise pixel dimensions

Fixed-ratio cropping controls proportions but not absolute size. If a form requires an image of exactly 800×600 pixels, crop to 4:3 ratio first, then use the Resize Image tool to set the exact pixel dimensions. Cropping followed by resizing is the correct two-step workflow for meeting precise dimensional specifications.

Common situations where exact pixel dimensions matter:

• Government document portals — visa applications, ID renewals, and permit forms often specify exact pixel dimensions and file size limits simultaneously. Crop to ratio first, then resize to the required dimensions, then compress if a file size limit applies.
• E-commerce product images — Shopify, Amazon, and WooCommerce have image dimension requirements that determine how products display in search results and on product pages.
• Social media ad formats — ad platforms specify exact dimensions for creative assets, and images that do not match the spec are automatically cropped by the platform, often in ways that cut off important content.
• Print templates — photo book layouts, business card templates, and poster designs require images at specific dimensions and resolutions (typically 300 DPI for print) to avoid blurry output.

How browser-based image cropping works

When you load an image in a browser-based crop tool, the image is read into memory using the browser's File API — the same API used to read file contents when you drag a file onto a web page. The selection you draw is tracked as four numbers: x offset, y offset, width, and height, all in pixels relative to the original image at full resolution.

When you click Crop, the tool draws the original image onto an off-screen HTML5 canvas at its full resolution, then reads back only the pixel data within your selection using the canvas's drawImage() method with source coordinates. That data is written to a new canvas of the selection's dimensions, then converted to the output format (PNG or JPG) using canvas.toBlob(). The resulting file is offered as a download using a temporary object URL.

This entire sequence runs inside the browser's JavaScript engine. There is no HTTP request to any server during the crop operation — the selection, the pixel extraction, and the file encoding all happen locally. For a 5MB JPG, the full crop takes well under a second on a modern device.

Limits and what to expect

• Very large images — images above roughly 30–50 megapixels can cause memory pressure in some browsers. This is relevant for high-resolution DSLR images (24MP+). Most phone photos and web images are well within browser limits.
• Output format follows input — a JPG input produces a JPG output; PNG in, PNG out. To change format, use the Convert Image tool before or after cropping.
• JPEG re-encoding — cropping a JPG re-encodes the JPEG data, which introduces a small additional generation loss. For most purposes this is imperceptible. If you are making repeated edits to a master image, work from PNG to avoid accumulating losses.
• EXIF data — canvas-based processing strips EXIF metadata (camera settings, GPS coordinates, orientation data). If you need to preserve EXIF, use a desktop editor. If you want to strip EXIF for privacy before sharing a photo, the Remove Image EXIF tool handles that separately.
• Browser support — Chrome 90+, Firefox 90+, Safari 15+, Edge 90+.

Privacy: what happens to your image

Your image is loaded into browser memory and processed using the HTML5 Canvas API. Nothing is sent to a server at any point. The cropped result is generated entirely in memory and downloaded to your device via the browser's built-in download API.

For images that contain sensitive content — identifiable faces, documents with personal information, location-tagged photos — browser-based cropping means the content never leaves your device. A server-based crop tool, by contrast, receives your full original image at their servers, processes it there, and returns the cropped version. The original is stored on their infrastructure for some period defined in their privacy policy — typically a few hours, in practice unverifiable.

The difference is structural, not just a matter of trust: a browser-based tool is technically incapable of uploading your file. Open the Network tab in DevTools while using any tool on this site and you will see zero upload activity during file processing.