Abstract
The k-means clustering algorithm is a commonly used algorithm for palette design. If an adequate initial palette is selected, a good quality reconstructed image of a compressed colour image can be achieved. The major problem is that a great deal of computational cost is consumed. To accelerate the k-means clustering algorithm, two test conditions are employed in the proposed algorithm. From the experimental results, it is found that the proposed algorithm significantly cuts down the computational cost of the k-means clustering algorithm without incurring any extra distortion.