Blur softens local detail by mixing nearby pixels. It can reduce visual distraction or create a background effect, but it is not a dependable way to certify that sensitive information has been removed.
Gaussian Image Blur applies a bounded number of Gaussian blur passes. Each additional pass changes the image further, so the result depends on both the source and the pass count.
Compare one pass with several
Begin with one pass and inspect a sharp edge. Increase the count when you need more softness. A tiny text label and a large face will respond differently because their detail exists at different scales.
Dimensions stay the same. The tool does not lower the file's resolution or select a particular background area automatically. It blurs the image as a whole.
A batch receives the same number of passes. That does not mean every image will look equally blurred when one is much larger or has finer detail than another.
Do not confuse an effect with redaction
Blurring a name, account number or face may leave recognisable information or allow someone to infer it from context. If information must be removed, use a suitable redaction process and verify the final exported pixels.
This tool does not assess whether a blur is strong enough for a privacy requirement. It simply applies the selected filter.
Keep an original when experimenting with the effect. A blurred download has already lost local detail, and another pass from that copy cannot restore it.
Review the output where it will actually be used. A preview may look soft at one zoom level and reveal more detail at another. Use blur for its visual purpose without claiming that it makes the contents unreadable to every observer.