What are PRNU and DSNU?

Photo Response Non-Uniformity (PRNU) and Dark Signal Non-Uniformity (DSNU) describe two different types of pixel-to-pixel variation in an image sensor. Both can contribute to fixed-pattern noise and variations in pixel brightness.

What is Photo Response Non-Uniformity (PRNU)?

Photo Response Non-Uniformity (PRNU) describes differences in the sensitivity of individual pixels to light.

Due to microscopic process variations during sensor manufacturing, pixels do not respond to exactly the same amount of incoming light in exactly the same way. When the sensor is exposed to a uniform light field, some pixels therefore produce slightly higher or lower signal values than their neighbors.

This can appear as a subtle fixed-pattern structure in a flat-field image, sometimes resembling a faint checkerboard or grid pattern.

PRNU becomes more noticeable as the amount of illumination increases because it is related to the pixel's response to light.

In simple terms:
PRNU means that pixels respond slightly differently to the same amount of light.

What is Dark Signal Non-Uniformity (DSNU)?

Dark Signal Non-Uniformity (DSNU) describes differences between pixels when no light reaches the sensor.

Even in complete darkness, pixels generate a small amount of signal due to dark current and other sensor-related effects. The amount is not exactly the same for every pixel. This pixel-to-pixel variation produces spatial non-uniformity in a dark image.

DSNU is therefore typically observed by capturing an image with the sensor covered or otherwise completely shielded from light. The resulting pattern is often referred to as fixed-pattern noise.

DSNU generally becomes more significant with longer exposure times and higher sensor temperatures, because dark current increases with temperature and exposure time.

In simple terms:
DSNU means that pixels produce slightly different signal levels even when no light reaches them.

What is the difference between PRNU and DSNU?

	PRNU	DSNU
Full name	Photo Response Non-Uniformity	Dark Signal Non-Uniformity
Occurs	Under illumination	In darkness
Main cause	Differences in pixel sensitivity	Differences in dark current / dark signal
Typical test	Uniform flat-field image	Dark image
Depends on	Amount of incident light	Exposure time and temperature
Typical appearance	Pixel-to-pixel brightness variations / fixed pattern	Pixel-to-pixel dark-level variations / fixed pattern

Both effects are inherent characteristics of image sensors. Their magnitude depends on the sensor technology, operating conditions and individual sensor characteristics.

Can PRNU and DSNU be corrected?

Yes. Both PRNU and DSNU can be reduced using pixel-level correction techniques, commonly referred to as flat-field correction, shading correction or fixed-pattern-noise correction.

A typical calibration process uses:

• a uniformly illuminated image to characterize PRNU

• a dark image to characterize DSNU

The resulting correction data can then be applied to subsequent images to compensate for pixel-to-pixel variations.

The availability and implementation of such corrections depends on the image sensor, camera model and image-processing pipeline.

Are PRNU and DSNU the same as image noise?

Not exactly.

PRNU and DSNU describe spatial non-uniformity between pixels. Because the resulting pattern is related to individual pixels, it can remain relatively consistent from frame to frame and is therefore often considered a form of fixed-pattern noise.

This differs from temporal or random noise, where the pixel value varies randomly from one exposure to another.

In practical terms:

PRNU / DSNU → spatial variation
Temporal noise → frame-to-frame variation