Title :
An evolutionary approach to color image quantization
Author :
Freisleben, Bernd ; Schrader, Andreas
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Siegen Univ., Germany
Abstract :
In order to visualize true color images on graphic displays with reduced color resolution, a color quantization process is required. Since color quantization is an NP-hard optimization problem, several suboptimal heuristic approaches, with quite different objectives and results, have been proposed. We present a new hybrid approach in which an evolutionary algorithm is combined with a well-known local search heuristic. The superiority of the proposed approach to other strategies used in color quantization is demonstrated by presenting results for some test images
Keywords :
data compression; genetic algorithms; heuristic programming; image coding; image colour analysis; image resolution; search problems; NP-hard optimization problem; color image quantization; color image visualization; evolutionary approach; graphic displays; hybrid approach; local search heuristic; reduced color resolution; suboptimal heuristic approaches; test images; Color; Evolutionary computation; Hardware; Image coding; Image resolution; Image storage; Pixel; Quantization; Testing; Visualization;
Conference_Titel :
Evolutionary Computation, 1997., IEEE International Conference on
Conference_Location :
Indianapolis, IN
Print_ISBN :
0-7803-3949-5
DOI :
10.1109/ICEC.1997.592355