In rasters we use interpolation to determine the value of a pixel, based on its surrounding pixels. The main raster-based interpolation methods are nearest neighbour, bilinear, and bicubic interpolation. To determine the value of the centre pixel (bold), in nearest neighbour interpolation the value of the nearest original pixel is assigned, i.e. the value of the black pixel in this example. Note that the respective pixel centres, marked by small crosses, are always used for this process. In bilinear interpolation, a linear weighted average is calculated for the four nearest pixels in the original image. In bicubic interpolation a cubic weighted average of the values of 16 surrounding pixels (the black and all grey pixels) is calculated. Note that some software uses the terms “bilinear convolution” and “cubic convolution” instead of the terms introduced above.