Use a Gaussian Sampling operator on an image to reduce image noise or produce a less-pixelated image, depending on the needs of your vision application:
The Gaussian Sampling operator processes an input image with a Gaussian kernel that approximates a two-dimensional Gaussian distribution. By varying the size of this kernel, the effect of the smoothing can be lessened or increased:
The operator examines the input image for the grey value of each pixel and the pixels surrounding it, takes a fraction of the grey value of each pixel as specified by the kernel, adds these values together, and assigns this new value to the corresponding pixel in the output image:
Choose a smoothing value which corresponds to the size, in pixels, of the features you want to reduce in detail. The output image is always the same size as the input image.
See the following topics for more information: