A Color Image Quantization Algorithm Based on Fuzzy Kernel Clustering
Yu Ma · Journal of Henan University of Science & Technology · 2011
A color image quantization algorithm based on fuzzy kernel clustering was studied in this paper.Firstly,the original image was quantized by using the median-cut algorithm.Then,the initial clustering centers were determined adaptively on the relationship between the NBS distance and the color difference for human visual.Finally,the clustering center color values were modified by use of the fuzzy kernel clustering algorithm in the Munsell space in order to get the quantization effect.The simulation result shows that the presented algorithm has better quantization effect than the median-cut algorithm and fuzzy c-means algorithm in the same quantization number.