摘要
提出了针对复杂背景的从前下视角度航拍机场图像中自动检测机场跑道的方法;基于引力场的模型可以直接建立图像失配代价函数的梯度场;按照最优估计理论,只要在正确匹配位置的收敛域内,就可以沿最速下降路径找到正确匹配位置;基于计算机视觉和惯性导航的组合导航方案,通过借助惯性导航系统提供的机场预测位置可以减少Hough变换所需的时间,保证了实时性的要求;仿真试验表明这种方法对复杂背景、前下视角度的航拍机场图像能够快速准确地检测、定位机场主跑道。
This paper discusses the methods to detect the runway in airborne forward-high-angle-shot images with complex background automatically.We can build the gradient field of the mismatch cost function directly based on the model of the gravitational field.As long as in convergence domain of the correct matching position,we will find the steepest descent path according to the optimum estimation theory and then find the correct matching position.The Hough transformation can be used to find straight lines in pictures.By means of INS,the predicted positions of the lines can help us to save the required time.This ensures the requirement of real-time be met.The experiments show that the algorithms have good performance of detecting and locating airfield runways in forward-looking aerial images.
出处
《计算机测量与控制》
CSCD
北大核心
2010年第10期2259-2261,共3页
Computer Measurement &Control
关键词
组合导航
图像匹配
引力场模型
HOUGH变换
检测跑道
composite navigation
image matching
gravitational field model
Hough transformation
detection of airfield runways
