论文部分内容阅读
传统的遥感图像机场跑道自动目标检测由于仅提取灰度特征常产生过分割现象,本文采用灰度特征和纹理特征相结合的方法进一步提高跑道的检测精度。利用阈值对遥感图像进行初始分割,以定位感兴趣区域(ROI),再利用EM算法估计ROI区域训练样本,引入马尔可夫随机场(MRF)模型,分割机场跑道。实验表明MRF可以很好地描述空间连续性,可以达到精确检测机场跑道的目的。
The traditional remote sensing image of airport runway automatic target detection often results in over-segmentation due to the extraction of only the gray-level features. In this paper, the combination of gray-level features and texture features is used to further improve the detection accuracy of runways. Thresholding is used to segment the remote sensing image to locate the region of interest (ROI). EM algorithm is used to estimate the training samples in the ROI region. Markov random field (MRF) model is introduced to segment the runway. Experiments show that the MRF can well describe the spatial continuity and achieve the purpose of accurately detecting the airport runway.