NxN filter

Use the Process > Smoothing Filters > NxN... command as an average filter. With the NxN filter, you can e.g., reduce the statistic noise in an image.

The dialog box that is opened when you use an image processing operation is made up in the same way for every operation. Click here to switch to a description of this dialog box.

Parameters for the NxN filter

The Iterations value determines how often the filter is to be used on the image. Values of from 1 - 25 are possible. A value of 3 is preset. As the number of iterations ascends, the averaging grade increases. The number of iterations and the matrix size also determine the time required to complete the filter operation.

The Size value determines the size of the filter matrix. The preset matrix size is 10. If you choose to make your matrix very large, the filter will affect large image structures.

Exactly what happens?

In a small image section, the intensity values of all of the pixels will be averaged and assigned to the central pixel. The name of the filter refers to the size of the image area whose pixels’ intensity are averaged - the size of this image area comprises precisely NxN pixels.

When determining filter size "N" you should essentially keep the following in mind: the smaller the matrix, the finer the details you can edit. These details can include artifacts or interference. Larger matrices suppress these effects, but yield more blurred resulting images - mean value filters smooth out intensity peaks. The higher the number of iterations, the greater this smoothing effect will be. You may end up blurring edges.

1702 05032013