基于形态梯度运算的遥感图像边缘检测

针对常规线性边缘检测器处理遥感图象时细节丢失严重的缺点,介绍了数学形态学基本理论,讨论了数学形态学在边缘检测中的应用.形态学的灰度梯度运算是在经典形态变换基础上提出的一类非线性算子.对于结构元素的选取作了一定的说明.另外,还与传统线性算子的处理结果进行了比较.通过计算机对遥感图像的模拟实验表明：基于形态灰度梯度运算的遥感图像边缘检测方法,不但几何意义明确,易于构造,而且性能也优于传统检测算子,证实了该方法的可行性.

Edge Detection Method of Remote Sensing Images Based on Mathematical Morphology of Gradient

In this paper,the basic theory of the morphology is presented and the morphology applied to detecting edge is discussed,in the light of the classical inear edge detector lose RS image details severely.The morphology of Gray Gradient is a non-linear operations based on standard morphological operations.We state the choice of the structuring element and compare with the classical edgede detection operators .The experimental results show the edge detection method of RS Images based on mathematical morphology of gray gradient better than the traditional operators not only the geometrical properties but also the effects of the structuring element and approve the feasibility of this method.