A new adaptive color image quantization algorithm
Chunhui Liu · Journal of Liaoning Normal University · 2007
Color quantization is an elementary problem in image analysis and image processing.Based on the K-mean clustering,a new adaptive color image quantization algorithm is proposed.Firstly,the initial clustering center is identified according to the image characteristics,and all image pixels are clustered.Secondly,the clustering center color values are modified according to the local image smooth.Finally,the clusters are split and merged,and the quantized image is produced.Experimental results show that our color quantization algorithm has the advantage of less distortion,which performs better than that of the K-mean clustering color quantization scheme.