Image Noise Generator

Add film grain, Gaussian noise, or salt-and-pepper noise to images.

Tested tool guide Tested browser tools Checked August 16, 2026

What Image Noise Generator does and how it behaves

Image Noise Generator changes the pixel texture of a source image by applying one of three distinct noise styles. Film grain adds an irregular photographic-looking texture, Gaussian noise spreads mostly small random variations with fewer large ones, and salt-and-pepper noise creates scattered bright and dark impulses. Noise amount is not the same as blur or sharpening: it introduces variation rather than recovering detail. The exact speck pattern is not the meaningful result; the mode, density, and visual scale are what should be judged.

How the result is produced

What each mode changes

Film grain overlays or changes fine tonal texture to evoke grain, without claiming a specific film stock. Gaussian noise makes random pixel deviations whose values follow a bell-shaped distribution, so modest changes occur more often than extreme ones. Salt-and-pepper noise is impulse noise: selected pixels become very dark or very light instead of receiving only a small offset. The visual signatures therefore differ even at similar apparent strength.

Scale, density, and bounds

Noise exists on the source image's pixel grid. Fine grain may become almost invisible when the result is reduced, while isolated salt-and-pepper pixels may merge or soften during resampling. Increasing Gaussian variation also affects already dark and bright areas, where visible values have less room before reaching an endpoint. Compare modes at the intended viewing size and inspect both smooth areas and edges.

Good uses

Adding restrained film grain to a digitally clean portrait or illustration when the desired finish is photographic texture rather than blur, sharpening, or a color shift.
Producing Gaussian-noise variants of test images to see whether edge detection, OCR, segmentation, or denoising work remains stable as pixel values fluctuate.
Creating salt-and-pepper corruption samples with unmistakable isolated light and dark pixels for teaching impulse noise or testing a median-filter workflow.

Limits and checks

Film grain is a visual style, not a calibrated reproduction of a named emulsion, ISO speed, scanner, or camera sensor. Do not read a grain setting as a physical measurement.
Gaussian describes the distribution of the added variation, not a promise that the finished image histogram will be Gaussian. The source pixels and bounded channel range still shape that histogram.
Salt-and-pepper marks can destroy original pixel values. A visually acceptable preview does not mean text, faces, or thin lines remain usable after stronger settings or later resizing.

Common questions

Which noise mode should I choose?

Use film grain for an image-wide photographic texture. Use Gaussian noise for random variation around existing tones. Use salt-and-pepper noise for sparse, extreme dark and light pixel faults. Similar-looking strength settings do not make the modes equivalent, so select by the kind of corruption or texture you need, not by file size.

Can the generated image be treated as a realistic camera-noise sample?

Not necessarily. Gaussian noise is a useful mathematical model, and film grain or salt-and-pepper noise can provide recognizable visual patterns, but real camera noise can depend on exposure, sensor behavior, processing, and color channels. Use the result for appearance or controlled testing. Do not present it as a calibrated sensor measurement unless you have separately validated it against that sensor.

References and verification

The behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.