DBSCAN(Density-Based Spatial Clustering of Applications with Noise)是一个比较有代表性的基于密度的聚类算法。与划分和层次聚类方法不同，它将簇定义为密度相连的点的最大集合，能够把具有足够高密度的区域划分为簇，并可在噪声的空间数据库中发现任意形状的聚类。(DBSCAN is a representative density based clustering algorithm. Unlike the partition and hierarchical clustering method, it will be the largest collection of clusters is defined as the density connected points, to have sufficient division of high density clusters, clustering and discovery of arbitrary shape in spatial database in noise.)