基于遥感图像分块直线特征检测的机场跑道检测方法

针对遥感图像机场跑道检测问题，提出了一种基于图像分块直线特征检测的机场跑道检测方法。首先，针对遥感图像数据量大带来的计算处理问题，设计了基于直线分割检测子（LSD）的遥感图像分块直线特征检测环节；然后，在总结归纳机场跑道数学特性的基础上，对提取的直线特征进行平行线分组、直线生长、平行线合并，并以Radon变换为基础，找出候选机场跑道区域；最后，使用灰度统计信息并结合梯度方向直方图对候选区域进行处理，筛选出最终的机场道路区域。实验结果表明，在能够提取出有效直线特征的情况下，该方法可以对多类机场跑道进行有效定位。

Airport Runway Detection Based on Line Feature Detection of Partitioned Remote Sensing Images

A method of airport runway detection is proposed based on line feature detection of partitioned remote sensing images. First, in order to deal with the computational difficulties arising from the huge data size of remote sensing images which may be hundreds of megabytes or even gigabytes, overlapped partitions are carried out on these images, and line feature detection is performed on each part by introducing the line segment detector (LSD). Then, based on the mathematical features of airport runways which are summarized and listed, the extracted line features are divided into several parallel line groups according to angles, and for each group, after line growing and merging, candidate airport runway areas are obtained with the help of Radon transform. Finally, airport runways are located and confirmed based on the gray-scale statistical information of the candidate runways and the histograms of oriented gradients. The performance of the proposed approach is validated by experiments carried on a series of remote sensing images downloaded from Google Earth, from which effective line features can be extracted.