一种基于方向形态学的图像轮廓提取方法

提出了一种基于方向形态学的图像轮廓提取新方法，该方法首先通过形态学梯度边缘检测算子实现图像的边缘检测，处理的过程中考虑到了边缘的方向信息，从而能够更好地跟踪图像中主要的边缘，最后通过滚动膨胀正确的提取物体的轮廓信息，在结构元素移动的过程中引入了方向控制条件因子，减小了计算的冗余度，实验结果表明该方法能够实现自动边缘检测与轮廓提取，精度较高。

Boundary detection based on directional morphology

This paper proposed a boundary detection method based on directional morphology. Firstly we use the morphology gradient edge detector to find the edge of the image. In this process we consider the direction information, so the major edges can be tracked. Then we use roll dilation to detect the boundary of the object in the image and introduce a direction control factor in order to reduce the computation redundancy. Experiments results show that this method can realize the edge and boundary detection automatically and have high accuracy.