Color Quantization Using Coreset Sampling

摘要: Color quantization is an important operation with many applications in computer graphics and image processing analysis. Clustering algorithms have been extensively applied to this problem. However, despite its popularity as a general purpose clustering algorithm, k-means has not received much attention the colour literature because of high computational requirements sensitivity initialization. In paper, we propose novel color method based on algorithm. The proposed utilizes adaptive initialization, deterministic sub-sampling efficient coreset construction attain speed quality quantization. Experiments set benchmark images demonstrate be significantly faster than while delivering nearly identical results.