Lanczos resampling

Lanczos resampling is a method for changing the sampling rate of a digital signal or for shifting it by a fraction of the sampling interval, and the same formula can also act as a low-pass filter. Each sample of the input is replaced by a translated and scaled copy of the Lanczos kernel, which is a sinc function windowed by the central lobe of a second, longer sinc function, and the sum of those copies is then evaluated at the points where new values are wanted. Because it interpolates smoothly between samples it is often applied to multivariate interpolation, for example to resize or rotate a digital image, and it has been considered the best compromise among several simple filters for that purpose. The filter was devised by Claude Duchon, who named it after Cornelius Lanczos because its construction uses the sigma approximation, a technique created by Lanczos.