高空遥感图像空间特征信息三维虚拟重建仿真

Three-Dimensional Virtual Reconstruction Simulation of Spatial Feature Information of High-Altitude Remote Sensing Image

卫星采集的高空遥感图像容易出现模糊区域,需要对高空遥感图像空间特征信息进行三维重建。传统的特征重建方法无法获取较好的效果,提出基于序列图像的特征重建方法。使用SIFT算法提取匹配空间目标图像序列特征,采用改进SFM方法重建并优化稀疏点云和稠密点云,以此为依据,按照三维点云数据信息完成目标表面三维虚拟重建。结合层次分析法与纹理映射关系模型,构建多层次纹理映射模型,采用傅里叶变换思想计算物体特征点间的空间位置关系,优化目标表面三维虚拟重建结果。实验结果表明:所提方法重建高空遥感图像效果真实、细致,对噪声具有较高的鲁棒性。

The fuzzy parts often occur in high-altitude remote sensing image collected by satellite.It is necessary to reconstruct the spatial feature information of high-altitude remote sensing image.The traditional feature reconstruction method cannot get a good result.Therefore,this paper focuses on a method to reconstruct the features of sequence images.Firstly,SIFT algorithm was used to extract the image sequence features matching the spatial targets.Secondly,the improved SFM method was used to reconstruct and optimize the sparse point cloud and dense point cloud.According to 3D point cloud data information,the 3D virtual reconstruction of target surface was completed.On this basis,the analytic hierarchy process was combined with the texture mapping relation model,and the multi-level texture mapping model was constructed.In addition,Fourier transform was used to calculate the spatial position relationship among the feature points of object,and thus to optimize 3D virtual reconstruction results of target surface.Experimental results show that the proposed method is real and meticulous on reconstruction of high-altitude remote sensing image.Meanwhile,this method has higher robustness to noise.