MaskErosion<T>

Models & Types · AiDotNet.Diffusion.MaskUtilities

Erodes a mask by shrinking masked regions, removing thin protrusions and small artifacts.

Implements: IDataTransformer<T, Tensor<T>, Tensor<T>>

For Beginners

Erosion makes the white (masked) area smaller. If your mask has small white specks or thin white lines, erosion removes them. It’s like “peeling” a layer off the edge of the masked region.

How It Works

Morphological erosion shrinks white (masked) regions by taking the minimum value in a local neighborhood. This removes small noise, thin connections, and smooths mask boundaries inward.

Constructors

Constructor Summary
MaskErosion(int kernelSize) Initializes a new instance of the MaskErosion class.

Properties

Property Summary
bool IsFitted { get; }
bool SupportsInverseTransform { get; }
int[] ColumnIndices { get; }

Methods

Method Summary
Tensor<T> Apply(Tensor<T> mask) Applies morphological erosion to a mask tensor.
Tensor<T> FitTransform(Tensor<T> data)
Tensor<T> InverseTransform(Tensor<T> data)
Tensor<T> Transform(Tensor<T> data)
string[] GetFeatureNamesOut(string[] inputFeatureNames)
void Fit(Tensor<T> data)