Gaussian Matrix Kernels

order

1

Image processing operations may perform poorly due to noise. Such operations are resampling an image to change its size, detecting edges or other features, analyzing texture, and so on. Smoothing by convolving the image with Gaussian kernels (or filters) is commonly used to correct for the effect of noise because of the interesting properties of these kernels: separability, associativity, and scale representation.

The image at the

th

i

row and

th

j

column is the impulse response of the filter

(i,j)

G

(x,y)

, where

G

is the Gaussian kernel.