What is a bilinear interpolation; how is this different from a nearest neighbor interpolation?
Bilinear and nearest neighbor interpolations are used to determine the color values of pixels in result images. To define the color of a pixel in a particular row and column of a result image, RasTools must determine from what location in the source image to obtain the color information. To determine this location, RasTools uses a true rotation or general perspective transformation (optionally followed by a triangle-based rubbersheeting adjustment). Then, RasTools uses a nearest-neighbor interpolation or bilinear interpolation of pixels in the source image to determine the color values for the new pixel. The nearest neighbor result is simply the color value of the source image pixel at the nearest row and column corresponding to the location provided by the transformations. If the source image is in one-byte/pixel gray-level or three-byte/pixel color format, RasTools can use a bilinear interpolation to determine the color of the pixel at the result location.