Virtual Reconstruction of Multi View Scene with Outliers Deleted from Laser Point Cloud Data
When pure laser scanning technology is used for virtual reconstruction of real scenes,there are too few effective point cloud data,poor coordinate balance and more chaotic coordinates.In order to improve the accuracy of virtual reality reconstruction,a multi view scene virtual reconstruction method based on laser point cloud outlier deletion is designed.Laser point cloud data of multi view scene are extracted,including corner feature extraction of laser data and corner feature extraction of visual data,the outlier threshold is set,the space length contained in the non outlier dataset in the 3D coordinate axis is calculated,and the outlier coordinate grid structure is set.The average distance between the point cloud data before and after the removal of outliers is calculated by extracting values from the point laser cloud data to complete the removal of useless data and realize virtual reconstruction.The experimental results show that the average distance difference before and after removal is significant,which shows that the removal of outliers has a significant effect on virtual reality reconstruction technology.In the virtual reality reconstruction scene obtained by using the proposed method,the obvious room contour can be obtained,and the reconstruction image of large objects is also very clear.
laser point cloudmachine visionmulti view scenevirtual reality reconstruction technologyoutliersfeature extraction
国家科技期刊平台
NSTL
万方数据
