论文部分内容阅读
雷达图像的线条特征提取算法一般分为三步:首先作图像预处理,然后采用特定的边缘检测算子提取边缘点,最后形成各种有意义的线条特征。本文采用两步检测算子来直接检测边缘点,不需要对图像进行预处理。首先用比率线探测器对SAR图像进行线特征的提取,然后用互相关探测器进行提取,最后融合二者的结果。本文对该算法进行仿真,证实了该算法能够很好的提取图像的线特征。
Radar image feature extraction algorithm is generally divided into three steps: first for image preprocessing, and then use a specific edge detection operator to extract edge points, and finally the formation of a variety of meaningful line features. In this paper, the two-step detection operator is used to directly detect edge points without preprocessing the image. First of all, we use the ratio line detector to extract the line features of the SAR images, then extract them with the cross-correlation detector and finally combine the results of the two. In this paper, we simulate the algorithm and prove that the algorithm can extract the line features of the image well.