基于简单线性迭代聚类和快速最近邻区域合并的图像分割算法

Image segmentation algorithm based on SLIC and fast nearest neighbor region merging

针对传统区域邻接图在描述数据结构时,搜索全局最优解难的问题,提出了一种基于简单线性迭代聚类(SLIC)和快速最近邻区域合并的图像分割算法。该方法在区域邻接图基础上引入了最近邻接图来优化全局搜索,首先用SLIC超像素算法将图像分割成小区域,利用区域邻接图(RAG)和最近邻接图(NNG)的邻接表数据结构来描述区域之间的关系,然后计算每个待合并区域与其所有邻接区域之间的不相似度函数值,最后合并不相似度值最小的区域。实验结果表明:本文方法能较好地将最相似的区域进行合并,与传统的区域合并算法相比,降低了合并计算的复杂度,大幅度提高了区域合并的准确性。

It is difficult to search the global optimal solution when traditional region adjacency graph is used to describe the data structure.To solve this problem,an image segmentation algorithm based on SLIC and fast nearest neighbor region merging is proposed.This method introduces the nearest neighbor graph to optimize the global search based on the region adjacency graph.First,an image is divided into small regions by SLIC superpixel algorithm,in which the relation between regions is described by the adjacency table structure of the Region Adjacency Graph (RAG)and the Nearest Neighbor Graph (NNG).Then,the value of the dissimilarity function is calculated between each region to be merged with all of its adjacent regions.Finally,the region with the least similarity is merged.The experimental results show that the proposed algorithm can combine the most similarregions,and reduce the complexity of the merging calculation compared with the traditional region merging algorithm,which greatly improves the accuracy of the regional merging.