论文部分内容阅读
针对有畸变且无地面控制点的无人机遥感影像,提出以分块方式提取图像中心区域特征点作为基准伪控制点对另一幅图像进行几何校正的算法。以两个图像中心连线的中垂线划分重叠区域为两块,选取一幅图像靠近中心点的块重叠区域内有效特征点为基准伪控制点,以第二幅图像上对应的特征点为待校正伪控制点,校正该块重叠区域;以类似的方法校正另一半重叠区域。试验结果证明,校正后地物点的坐标与基准影像上该地物点的坐标的几何畸变残差平均值比校正前大幅度减小,有明显的校正效果。
Aiming at the remote sensing image of UAV with distortion and without ground control points, this paper proposes an algorithm of segmenting the feature points of the image center region as the reference pseudo-control points to geometrically correct another image. The center of the two image centers connected by the vertical line is divided into two overlapping areas, select an image near the center of the block overlap area valid feature points as the base pseudo-control point to the second image corresponding feature points for The dummy control points are to be corrected, the overlap area of the block is corrected, and the other half of the overlap area is corrected in a similar way. Experimental results show that the average value of the geometric distortion residuals of the coordinates of the corrected feature points and the coordinates of the feature points on the reference image decreases significantly compared with that before the correction, which has a significant correction effect.