Utilities tool

Image cropper

Crop an image to precise dimensions or a ratio.

Your inputs

Processed locally. Choose Uploaded files to use your files; generated sample mode ignores uploads.

Your results

Ready when you are
Add your inputs or try an example.
Your results will appear here.

How it works

Crop an explicit rectangle from decoded image coordinates and export the selected pixels as PNG without stretching or resampling.

Read the source reference ↗

A few things to know

No automatic subject detection or drag selection. Coordinates must fit within the decoded oriented image. Limits: 25 MB total, 20 images, 16 million pixels per source and output, and 8,192 pixels per side. Files are flattened, metadata is not copied and animations become one frame. JPEG uses a white background for transparency.

A little context

When to use this tool

Select a precise image rectangle using source-pixel coordinates. Width and height define the output ratio without stretching the selected pixels.

  • Create a precise thumbnail.
  • Remove image edges.
  • Crop to known dimensions.

How to use image cropper

  1. Prepare the inputs

    Choose an uploaded file or the generated sample.

  2. Choose the settings

    Enter dimensions using the source’s decoded pixel coordinates.

  3. Review and export

    Inspect the processed preview and download the flattened output.

Try the numbers

Try the supplied example

Inputs

source: Generated sample; x: 100; y: 100; width: 600; height: 400

Result

The example extracts a 600 × 400 rectangle starting at x=100, y=100 from the sample.

What it means

Check the preview to ensure the crop retains the intended content.

Common mistakes to avoid

Use decoded orientation when choosing coordinates

Use decoded orientation when choosing coordinates.

A crop removes everything outside the rectangle

A crop removes everything outside the rectangle.

Useful answers

Frequently asked questions

Definitions, edge cases, and practical next steps for this tool. Search all tool FAQs ↗

Which coordinates are used?

Pixels in the decoded, orientation-adjusted source image.

Can the rectangle extend beyond the source?

No. All edges must stay inside the image.

Is the output stretched?

No. The selected pixels are copied at the same dimensions.

Can I choose a ratio?

Enter width and height with the desired ratio; the aspect-ratio calculator can help.

What format is downloaded?

A flattened PNG, preserving transparency where the browser supports it.

Can I drag a crop box?

This version uses explicit numeric coordinates and a processed preview.

Are my inputs uploaded or stored?

Processing happens locally in your browser. This tool does not upload inputs. Downloads stay on your device; sharing or opening them in another service is your choice.

What should I check before using the output?

No automatic subject detection or drag selection. Coordinates must fit within the decoded oriented image. Limits: 25 MB total, 20 images, 16 million pixels per source and output, and 8,192 pixels per side. Files are flattened, metadata is not copied and animations become one frame. JPEG uses a white background for transparency.

Explore the topic in the blog

Sources and method

Use the references below to check definitions and platform guidance. The model’s scope and limits are explained above. Read our editorial standards.

Help

How can we help?

Find instructions, report a problem, or leave a message.

Read this tool’s FAQs ↗

Guides for this tool

Choose what the frame should preserve.
Common questions

Where do I start? Try an example, then replace its values with yours.

Why did my upload fail? Check the field error, accepted format and file-size limit. Start with a small sanitized sample.

Where are my results? Review the result panel, then use its copy or download controls. If a calculation fails, check the error beside the input.

Message us

Chat opens Tawk.to. It processes your messages and connection data. Do not paste private inputs or file contents. Availability and response times vary.