Image stack rendering modes

Paskutinį kartą atnaujinta 2026-10-29

Find all the stack modes with formulas and use cases for noise reduction and object removal in Adobe Photoshop on desktop.

All stack modes operate on a per-channel basis and calculate values using only non-transparent pixels in the stack. For example, the Maximum mode returns the maximum red, green, and blue channel values for a pixel cross-section and merges them into one composite pixel value in the rendered image.

Available stack modes

Mode

Result

Use cases

Entropy

entropy = - sum( (probability of value) * log2( probability of value) )Probability of value = (number of occurrences of value) / (total number of non-transparent pixels)

Defines the minimum bits needed to encode information in a set (binary or zero-order entropy).

Kurtosis

kurtosis = ( sum( (value - mean)⁴ ) over non-transparent pixels ) / ( ( number of non-transparent pixels - 1 ) * (standard deviation)⁴ )

Measures distribution shape. Returns 3.0 for normal distribution, above 3 for peaked, below 3 for flat.

Maximum

The maximum channel values for all non-transparent pixels

Preserves the brightest values across the stack.

Mean

The mean channel values for all non-transparent pixels

Reduces noise by averaging pixel values to smooth variations.

Median

The median channel values for all non-transparent pixels

Reduces noise and removes transient objects by selecting middle values that filter out temporary elements.

Minimum

The minimum channel values for all non-transparent pixels

Preserves the darkest values across the stack.

Range

Maximum minus the minimum of the non-transparent pixel values

Shows value spread for analyzing variation.

Skewness

skewness = (sum( (value - mean)³ ) over non-transparent pixels ) / ( ( number of non-transparent pixels - 1 ) * (standard deviation)³ )

Measures distribution symmetry around the mean.

Standard Deviation

standard deviation = Square Root(variance)

Quantifies how much values vary from the mean.

Summation

The sum channel values for all non-transparent pixels

Accumulates all values for specialized analysis.

Variance

variance = (sum( (value-mean)² ) over non-transparent pixels ) / ( number of non-transparent pixels - 1)

Measures how far values spread from their average.

Common workflows

Reducing image noise: Apply Mean for general noise reduction or Median when the scene includes occasional unwanted elements. Use Mean for purely random noise; choose Median when you need both noise reduction and the ability to filter transient objects.

Removing unwanted objects: Use the Median stack mode when you have a series of frames showing transient objects (people walking through a scene, passing vehicles) that you want to eliminate. The median calculation automatically excludes values that appear in only some frames.