一种基于模糊熵和FKCN的边缘检测方法

根据邻域内的灰度分布是有序的、具有方向性以及灰度实质具有结构性三种图像边缘的基本特征,通过引入模糊熵,构造出了一组基于模糊熵的信息测度分量来定量描述图像的这些边缘特征.利用训练样本获取该组分量,并组成一特征向量对模糊Kohonen聚类网络（FKCN）进行训练.用训练好的FKCN聚类网络直接提取其它图像的边缘.该方法无需确定阈值,对弱边界检测较敏感,在特征的选取上充分地考虑了边缘和噪声的本质区别,具有优异的抗噪性能.

A Method of Edge Detection Based on Fuzzy Entropy and FKCN

The three natural characters of image edge are that the gray intensity distribution in neighborhood of image is ordered,directive,and structurized.Through quoting fuzzy entropy,the authors construct three information measures fuzzy entropy-based to describe the natural characters of image edge.Three component vectors for describing three natural characters of image edge are gotten via trained image samples.Firstly,a fuzzy Kohonen clustering network(FKCN) is trained with some feature vector consist of the three component vectors,and then the edge points in a new image are directly extracted by using the trained FKCN.The method does not need any threshold;more sensitive for weak edge detection;has better anti-noise performance since the influence of noise is adequately considered when the feature vector is selected.