结合密度峰聚类的K均值图像分割算法

K-means Image Segmentation Algorithm Combined with Density Peak Clustering

针对K均值聚类算法在图像分割应用中的不足,结合密度峰聚类算法对原有算法进行改进,得到了一种图像分割效果较好的改进K均值算法。K均值算法需要人工指定聚类中心数目,并且聚类中心的随机初始化对最终的图像分割结果有很大影响。针对以上缺点,对K均值算法进行改进,通过密度峰聚类算法自动确定了图像分割的聚类中心数目和较为准确的初始聚类中心。为了衡量色差在人眼中的感知情况,在算法中引入了NBS距离作为距离测度。实验结果表明,改进后的图像分割算法在分割图像时具有稳定的性能和较好的效果。

Aiming at the deficiency of K-means clustering algorithm in image segmentation,an improved K-means algorithm with better image segmentation effect was proposed by combining density peak clustering algorithm to improve the original algorithm.The K-means algorithm needs to specify the number of clustering centers manually,and the initialization of clustering centers has a great impact on the final image segmentation results.In view of the above shortcomings,the K mean algorithm was improved.The number of clustering centers and the more accurate initial clustering centers for image segmentation were automatically determined by density peak algorithm.In order to measure the perception of chromatic aberration in human eyes,NBS distance was introduced into the algorithm as distance measure.Experimental results show that the improved image segmentation algorithm has stable performance and good effect in segmentation of images.