摘要
道路是遥感图像的重要特征。提出一种从遥感图像中检测道路目标的方法。首先,进行图像预处理,提取边缘,然后进行图像分块并利用Hough变换进行道路检测。分析了Hough变换的特点,并研究了Hough变换域极值点参数的特点,改进了Hough变换在道路目标检测中的应用。该方法不仅快速地检测机场等严格的线状道路目标,而且对具有一定曲率的道路目标同样有效。
Roads are important features of remote sensing images. In this paper,we propose an algorithm for detecting road from a remote sensing image. Firstly,a pre- processing method is used to detect the edges. Then,we divide the image obtained and detect road by Hough transform. From analyzing the characters of Hough transform and researching on the parameters of Hough transform, we improve the Hough transform in road detection. This method is not only efficient in airdrome detection but also available for the road detection with curvature.
出处
《现代电子技术》
2006年第6期125-127,共3页
Modern Electronics Technique
基金
空间微波技术国家级重点实验室基金项目资助
关键词
遥感图像
HOUGH变换
道路检测
目标识别
remote sensing image
Hough transform
road detection
object recognition
