基于数学形态学和Hough变换的复杂形态线状地物提取方法

为改善以Hough变换提取遥感影像中弯弧线状地物的准确性,提出了一种新算法M&H (Morphology and Hough),它将Hough变换、数学形态学、非线性拟合等方法串行递进使用,以实现复杂形态线状地物自动提取和直接生成道路矢量图的目标。经适量实例的Matlab仿真测试,证明了M&H的递进噪声去除法优于单次数学形态学降噪法,对含弯弧道路的提取性能也优于标准Hough变换。精度检验结果表明:M&H方法比常规Hough变换提取道路的欠提取率下降了约18.9%。

Complex Linear Topographic Features Extraction Using Mathematical Morphology and Hough Transform

The new algorithm called M&H (Morphology and Hough) is aimed at extracting complicated curved roads.H&M extends the theory and techniques of Hough transform,mathematical morphology and nonlinear regression.Eventually,H&M realizes the goal of automatically extracting complicated linear features and instantly generating the vector graphics of the roads.After several tests using MATLAB,the results show that M&H has advantage in noise reduction than traditional morphological method,and is much better than Hough transform at curved road extraction.The data of accuracy tests shows that M&H decreases the rate of extraction losses by 18.9%,which compared with simply using standard Hough transform.