快速均值漂移图像分割算法研究

Mean shift算法是一种搜索与样本点分布最接近模式的非参数统计方法.但它是一种迭代统计方法,要保证较高的数值计算精度需要较多的迭代次数,耗费较长的计算时间.为克服这一缺点,提出快速均值漂移图像分割算法.该算法在每次迭代时以前一次的聚类中心集合T动态地更新样本集S,并通过使用直方图缩小样本点的搜索范围进一步加快算法的收敛速度.实验结果表明该方法在保证图像分割质量的同时具有较快的收敛速度.

Research on Fast Mean Shift Algorithm for Image Segmentation

Mean shift algorithm is a nonparametric statistical method for seeking the nearest mode of a point sample distribution.However,it is a statistical iteration scheme which wastes more time to calculate and needs more iteration to ensure higher numerical accuracy.In order to overcome its disadvantage,fast mean shift algorithm for image segmentation is proposed.The sample set S is updated iteratively based on the set of cluster centers T computed in the previous step and the search band of sample point is reduced by means of histogram when using the fast mean shift algorithm for image segmentation,which can accelerate the velocity of convergence.Experimental results show that new algorithm have faster velocity of convergence and can get the better quality of image segmentation at the same time.