摘要
以西宁市某城镇区域为例,首先利用大疆DJI M600Pro无人机搭载的DG3 Pros倾斜摄影系统采集影像数据;再利用PhotoScan影像匹配模块处理无人机倾斜影像获取的多视影像数据同名点坐标;然后利用GodWork的AT模块进行引入曝光延迟的光束法区域网平差,并通过分块平差与整体一致性平差实现大区域多视倾斜影像数据的整体平差;最后利用Smart3D完成多视影像密集匹配和纹理自动映射,生成高精度的三维模型。通过控制数据对空三加密、三维立体模型进行了精度验证,结果表明,其精度能满足大比例尺测图要求。
Taking a town area of Xining City for example,we used DG3 Pros oblique photogrammetry system carried by Dajiang DJI M600 Pro UAV to collect image data,and used Photo Scan image matching module to process the coordinates of same name points of multi-view image data obtained by the oblique UAV images.And then,we used AT module of God Work to adjust the area network by introducing exposure delay,and realized the overall adjustment of large area multi-view oblique image data by block adjustment and overall consistency adjustment.Finally,we used Smart3 D to complete dense matching of multi-view images and automatic texture mapping,and generated high-precision 3 D model.We verified the accuracy of aerial triangulation and 3 D model through control data.
出处
《地理空间信息》
2021年第7期31-35,I0005,共6页
Geospatial Information
