Indexing method and device for large-scale point cloud data, terminal equipment and medium

Abstract

The invention discloses an indexing method and device for large-scale point cloud data, terminal equipment and a medium. The indexing method comprises the following steps: reading point cloud data from a local hard disk, wherein the point cloud data is that a spatial feature data structure of a target file is stored in a local hard disk by taking a file name form as an entry of an index; calculating a minimum bounding box vertex of the point cloud data to obtain X-axis, Y-axis and Z-axis coordinates of the maximum bounding box and the minimum bounding box; constructing a file index table according to the coordinates; and directionally extracting target point cloud data from the file index table, the target point cloud data being that the processing pressure of a computer memory can be reduced, and the indexing efficiency of large-scale point cloud data is improved.

Description

technical field [0001] The invention relates to the technical field of computer data structures, in particular to an indexing method, device, terminal equipment and media for large-scale point cloud data. Background technique [0002] Point cloud data has the characteristics of large data volume (massiveness), fine data expression (high spatial resolution), and no topological relationship (scatteredness) between spatial three-dimensional points. In subsequent data processing, frequent neighborhood searches are required. , so the octree structure or K-D tree data structure is often used for data organization to realize the indexing of point cloud data. [0003] The octree data structure is in the form of continuously cutting the point cloud data into 8 sub-cubes according to the cube. When the computer memory is large enough, the super-large point cloud data can be indexed by using the octree data structure. However, each index needs to re-read the point cloud data into the...

AI Technical Summary

Problems solved by technology

[0004] The above two methods are to establish a tree structure based on the point cloud data in the memory, which requires high computer memory.

In addition, for ultra-large point cloud data, each index needs to read a large amount of point cloud data into the memory to establish a tree structure before indexing, and the indexing efficiency is low

Images