A dynamic niching clustering algorithm based on individual-connectedness and its application to color image segmentation
作者：
Highlights：
• The DNIC aims to automatically evolve the number of clusters and the cluster centers.
• A simpler representation is adopted that each individual represents a single center.
• An algorithm is proposed to select the optimal number of the neighbor automatically.
• An individual-connectedness algorithm is proposed to dynamic identify the niches.
• The dynamic niching is accomplished without assuming any a priori knowledge.
摘要
Highlights•The DNIC aims to automatically evolve the number of clusters and the cluster centers.•A simpler representation is adopted that each individual represents a single center.•An algorithm is proposed to select the optimal number of the neighbor automatically.•An individual-connectedness algorithm is proposed to dynamic identify the niches.•The dynamic niching is accomplished without assuming any a priori knowledge.
论文关键词：Clustering,Genetic algorithms,Niching,Connected individual,k-distance neighborhood,Image segmentation
论文评审过程：Received 15 September 2014, Revised 29 February 2016, Accepted 9 May 2016, Available online 21 May 2016, Version of Record 11 June 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.05.008
