基于变尺度形态学的遥感图像边缘检测算法

针对遥感图像的噪声问题,在数学形态学的基础上提出了一种利用结构元素尺度变换的边缘检测算法。首先利用腐蚀运算对图像降低噪声,然后利用膨胀运算填补腐蚀产生的空穴,在这个过程中采用不同尺度的结构元素,最后通过检测算子得到图像的边缘图。实验结果表明,与传统的边缘检测算法相比,该算法不仅具有较强的边缘提取能力,而且通过合理选择结构元素的尺度能明显降低噪声对检测结果的影响,准确检测出大多数遥感图像的边缘。

Edge detection algorithm for remote sensing image based on scale transformable morphology

Remote sensing images usually contain much noise which affects their edge detection accuracy. For this problem, a new edge detection algorithm based on mathematical morphology is presented in this paper. Firstly, the image is eroded to reduce its noise, and then inflated to fill in the holes by erosion with different construction element. Finally the edge picture of the image is obtained through the operator proposed. The experimental results show that the algorithm proposed in this paper can considerably reduce noise in most remote sensing images and accurately extract their edge pictures. Compared with the conventional algorithms, it is more feasible for noisy image processing.