基于角度搜索和距离判定的边缘检测算法

在离散数据处理中,边缘检测算法的应用非常广泛。针对当前边缘检测算法存在的问题,如数据量的大小、数据的几何特性、密度的集中程度、距离跨度等,提出了一种基于角度搜索和距离判定的边缘检测算法,详细考虑了边界的多情况、凹边形和凸边形的几何特性。首先建立四个点的边界凸壳,然后采用角度判断的方法进行凸壳内缩,最后进行距离判断的二次内缩检验。该方法简洁、高效、便于理解与实现,并且可以针对各种情况的离散数据。采用乐山市城区的商业圈聚类分析数据进行测试,该离散数据体现了高密度、低密度、数据量大、距离跨度大等特点,通过实验证明了该算法边界检测效果较好,异常数据小,能够准确甄别出边界区域数据。同时也证明了该算法的通用性、简洁性和高效性。

Edge Detection Algorithm Based on Angle Search and Distance Judgment

Edge detection algorithms are widely used in discrete data processing.In view of the problems existing in the current edge detection algorithms,such as the size of data volume,geometric characteristics of data,concentration degree of density,distance span and so on,we propose an edge detection algorithm based on angle search and distance determination,considering the multi-case of boundary,geometric characteristics of concave and convex in detail.First,the boundary of the four point convex hull is established,then the method of angle judgment within convex hull is adopted,and finally the distance judgment within the secondary shrinkage test is carried out.This method,which is simple,efficient,and easy to understand and implement,can be used for all kinds of discrete data.The commercial cluster analysis data of Leshan city are tested.The discrete data reflects the characteristics of high density,low density,large data volume,and large distance span and so on.Experiment shows that the algorithm is effective in boundary detection,with small abnormal data,and can accurately identify the boundary data.At the same time,its generality,simplicity and efficiency are also proved.