Choose image dimensions, compression, and format

Understand when to crop, resize, compress, or convert an image for a real upload requirement.

An image can have the correct width and still be too large to upload. It can also have a small file size while looking blurry. Pixel dimensions, composition, encoding quality, and file format are separate choices. Handle them in that order so you can see what each change does.

Read the destination’s requirement first

Suppose a form requires a square image at least 600 × 600 pixels and no more than 1 MB. “Square” is a composition requirement, “600 × 600” describes dimensions, and “1 MB” limits the encoded file size. You need to satisfy all three. Do not stretch a rectangular photo into a square just to match the numbers.

Crop for composition, then resize for dimensions

Use Crop Image to select a square region that keeps the subject. Then use Resize Image to set the required pixel dimensions. A 2400 × 1600 image reduced proportionally to width 1200 becomes 1200 × 800. Forcing both dimensions to 1200 changes the proportions and can make faces or logos look distorted.

Select a format for the content

JPG is often suitable for photographs. PNG is useful for transparent graphics and sharp screenshot edges. WebP is an option when the destination supports it. If you use PNG to JPG, inspect the background: JPG cannot preserve transparency. Keep the source PNG if you need a transparent version later.

Compress only as much as the task needs

Use Compress Image and compare the downloaded size with the limit. Inspect small lettering, fine textures, and the edges around the subject at normal viewing size and at 100% zoom. A smaller result is not useful if an ID number or product label becomes unreadable. Repeatedly recompressing the latest copy can accumulate loss, so return to the original when trying another setting.

Verify the exported file

Open the actual download, confirm its file type and dimensions, and test it in the destination. Browser canvas exports may discard metadata, animation, or color-profile information. Increasing the dimensions of a low-resolution image cannot recover missing detail. Keep your original for other uses and use the optimized copy for the specific task.