Gaussian Blur
A filter that convolves an image with a Gaussian kernel to produce smooth, natural-looking blur by attenuating high-frequency detail.
Gaussian blur applies a weighted convolution kernel shaped by the normal (Gaussian) distribution to suppress high-frequency components and produce a smooth defocusing effect. The standard deviation σ controls blur strength: σ = 1.0 affects roughly a 3-pixel radius, while σ = 3.0 extends to about 9 pixels.
A key advantage is separability: the 2D kernel can be decomposed into two 1D passes (horizontal then vertical), reducing computational complexity from O(n²) to O(2n). This makes large kernel sizes practical even on resource-constrained devices.
Applications include noise reduction, simulated depth-of-field, and preprocessing for edge detection. CSS filter: blur() applies Gaussian blur natively. In unsharp masking, the difference between the original and a Gaussian-blurred copy is used to enhance sharpness.