论文部分内容阅读
针对遥感影像具有丰富的细节和混合噪声,采用传统高斯拉普拉斯(LOG)算子难以有效地提取遥感影像地物边缘信息,文章提出一种结合LOG算子和大津法(Otsu法)的遥感影像地物边缘信息提取算法。此算法采用中值滤波对原始影像进行消噪,利用Otsu算法对滤波后影像进行自适应阈值分割得到目标影像,采用LOG算子对目标影像进行边缘信息提取。以安宁某区域的遥感影像为例,实验结果表明,提出的算法与传统LOG算子相比,具有更高的提取精度,能有效地提取出遥感影像的真实边缘、减少伪边缘。
In view of the rich details and the mixed noises of remote sensing images, it is difficult to extract edge information of remote sensing image by traditional Gaussian Laplace operator. In this paper, we present a new method combining LOG operator and Otsu method Remote sensing image feature edge information extraction algorithm. This algorithm uses median filtering to denoise the original image, uses Otsu algorithm to adaptively threshold the filtered image to get the target image, and uses LOG operator to extract the edge information of the target image. Taking the remote sensing image of a certain area as an example, the experimental results show that the proposed algorithm has higher extraction accuracy compared with the traditional LOG operator, and can effectively extract the real edge of remote sensing image and reduce the false edge.