边缘提取：边缘检测的基本思想首先是利用边缘增强算子，突出图像中的局部边缘，然后定义象素的“边缘强度”，通过设置阈值的方法提取边缘点集。由于噪声和模糊的存在，监测到的边界可能会变宽或在某点处发生间断。因此，边界检测包括两个基本内容：
（1）用边缘算子提取出反映灰度变化的边缘点集。
（2）在边缘点集合中剔除某些边界点或填补边界间断点，并将这些边缘连接成完整的线。(Edge extraction: the basic idea of edge detection is to use edge enhancement operator to highlight the local edge of image, and then define the edge intensity of pixels, and set the threshold method to extract the edge point set. Due to the presence of noise and blurring, the monitored boundaries may be widened or discontinuous at some point. Therefore, the boundary detection includes two basic contents:
(1) edge operators are used to extract the set of edge points that reflect the change of gray level.
(2) eliminate some boundary points in the set of edge points or fill the breakpoints between the boundaries, and connect these edges into a complete line.)