基于聚类算法的多态单模板掩模边缘检测算法

提出了一个能够动态估计邻域中边缘位置的基于聚类算法的多态单模板掩模边缘检测算法。该算法根据估计出来的边缘位置设置模板上相应位置的权值，可以获得更为准确的差分结果。同时根据边缘的连续性，排除不连续的伪边缘，以在获取较高分辨率的同时减低噪声。实验证明基于聚类算法的多态单模板掩模边缘检测算法具有良好时实性及较高的分辨率。

Polymorphisms single template masking for edge detection based on clustering algorithm

A polymorphisms single template masking for edge detection algorithm based on clustering was proposed, which could dynamically estimate the edge location of the neighborhood. The algorithm would set the weight parameters on the template by evaluating the position of the possible edge, which would give a more precise result. It made it possible to reduce the noise but keep a higher resolution by debarring the fake edge according to its continuity. Experimental results prove the effectiveness of this algorithm.