论文部分内容阅读
通过分析高分辨率光学遥感影像上的道路特征,提出了适用于有较多车辆遮蔽和两侧行树阴影干扰的T型模板匹配的道路提取算子。它融合了剖面匹配和模板匹配的优势,使用角度纹理特征进行初始道路点的精确定位,道路宽度、道路前进方向的精确量算,使用灰度最小二乘匹配进行最佳道路点的定位,可以很好地追踪高分辨率光学遥感影像上受车辆遮蔽和行树阴影等噪声干扰的带状道路。
By analyzing the road features of high resolution optical remote sensing images, a road extraction operator suitable for T-shape template matching with more vehicle shading and shading on both sides of the tree is proposed. It combines the advantages of profile matching and template matching, the use of angle texture features for the accurate positioning of the initial road point, the road width, the precise measurement of the road forward direction, the use of gray least squares matching for the best waypoint positioning, you can Good tracking of banded roads in high-resolution optical remote sensing images that are interfered with by vehicles such as shadowing and tree shadows.