Patent US 8,077,993

In an embodiment, a method includes receiving values of a first pixel and a second pixel of an image at an input processor element of an image signal processor and distributing fractional parts of a number of values from surrounding pixels to the first pixel and the second pixel non-serially using first and second accelerator units, respectively, of the image signal processor. The method further includes assigning output values to the first pixel and the second pixel based on the values of the first pixel and the second pixel and the fractional parts of the number of values from the surrounding pixels using an output processor element of the image signal processor and outputting the output values to an output device that has a color range that is less than a color range of the image using the output processor element.