Dave Mitchell's DigitalMicrograph™ Scripting Website

The Hanning Window filtering function creates a bell-shaped function with a value of unity in the centre and zero at the edges.

The bell-shaped Hanning function is used to create a filtering mask. This is typically applied when cross-correlating two images. The Hanning function will eliminate edge features and instead emphasise those in the centre. Images to be cross-correlated (eg for drift correction) are often first filtered with an edge-enhancing filter such as the Sobel filter, to improve the sensitivity of cross-correlation. The Hanning Window is then applied to these filtered images before cross-correlation, to eliminate any near-edge features from the analysis.