| 摘要 |
PROBLEM TO BE SOLVED: To provide an image processing apparatus and method for realizing quantization of a multi-value color image having two color components or more through error spread processing to have the number of gradations less than the original number of gradations by using a lookup table having a small memory capacity. SOLUTION: When data have color components 0, 1 configuring received pixels, an error caused in the quantization processing is added to respective data positions and the result is given to an output setting section 2. Each of one-dimensional LUTs receives the data as addresses, quantizes the data in a way of having a narrower range than that of the original range of the data and outputs the result as g0, g1. An N-dimensional LUT receives the data g0, g1, regarded to have a correlation of the received data having the two color components, and outputs the final quantization result. Since the size of the memory space of the N-dimensional LUT is controlled by the g0, g1, this means that a narrower range taken by the g0, g1 is better. On the other hand, a wider range of the g0, g1 provides a quantization result more suitable for forming an image with high quality. Then the g0, g1 are decided so that the image quality is minimized within a permissible range. COPYRIGHT: (C)2004,JPO
|
