论文部分内容阅读
遥感图像机场跑道边缘的提取是机场识别的主要方法。传统Hough变换在线段提取方面具有较高抗噪性,但用于遥感图像机场跑道提取时存在边缘定位性较差及弯曲跑道误检率高的问题。本文提出了图像空间多尺度Hough变换方法,提高了Hough变换在提取、检测线段时的定位能力;将传统Hough变换的对参数空间改进为相对参数空间,增强了Hough变换检测小线段的性能,并利用分段线段的连接,达到机场跑道边缘中直线段与曲线线段检测的目的。试验结果表明该方法在保持Hough变换高抗噪性的同时,可有效地检测复杂背景下遥感图像中的机场跑道边缘,并保证了边缘的连接性。
Remote Sensing Image Airport runway edge extraction is the main method of airport identification. The traditional Hough transform has high noise resistance in the extraction of the line segment, but it has the problems of poor edge locating ability and high false detection rate of the curved runway in the airport runway extraction of remote sensing images. In this paper, the multi-scale Hough transform in image space is proposed to improve the Hough transform’s ability to locate and extract line segments. The traditional Hough transform is improved to a relative parameter space, which enhances the performance of Hough transform to detect small segments The use of segment line connection to achieve the airport runway edge straight line and curved line detection purposes. The experimental results show that this method can effectively detect the edges of airport runways in remote sensing images and ensure the edge connectivity while maintaining the high noise immunity of Hough transform.